draft-ietf-ippm-model-based-metrics-13.txt   rfc8337.txt 
IP Performance Working Group M. Mathis Internet Engineering Task Force (IETF) M. Mathis
Internet-Draft Google, Inc Request for Comments: 8337 Google, Inc
Intended status: Experimental A. Morton Category: Experimental A. Morton
Expires: March 19, 2018 AT&T Labs ISSN: 2070-1721 AT&T Labs
September 15, 2017 March 2018
Model Based Metrics for Bulk Transport Capacity Model-Based Metrics for Bulk Transport Capacity
draft-ietf-ippm-model-based-metrics-13.txt
Abstract Abstract
We introduce a new class of Model Based Metrics designed to assess if This document introduces a new class of Model-Based Metrics designed
a complete Internet path can be expected to meet a predefined Target to assess if a complete Internet path can be expected to meet a
Transport Performance by applying a suite of IP diagnostic tests to predefined Target Transport Performance by applying a suite of IP
successive subpaths. The subpath-at-a-time tests can be robustly diagnostic tests to successive subpaths. The subpath-at-a-time tests
applied to critical infrastructure, such as network interconnections can be robustly applied to critical infrastructure, such as network
or even individual devices, to accurately detect if any part of the interconnections or even individual devices, to accurately detect if
infrastructure will prevent paths traversing it from meeting the any part of the infrastructure will prevent paths traversing it from
Target Transport Performance. meeting the Target Transport Performance.
Model Based Metrics rely on mathematical models to specify a Targeted Model-Based Metrics rely on mathematical models to specify a Targeted
Suite of IP Diagnostic tests, designed to assess whether common IP Diagnostic Suite, a set of IP diagnostic tests designed to assess
transport protocols can be expected to meet a predetermined Target whether common transport protocols can be expected to meet a
Transport Performance over an Internet path. predetermined Target Transport Performance over an Internet path.
For Bulk Transport Capacity the IP diagnostics are built using test For Bulk Transport Capacity, the IP diagnostics are built using test
streams and statistical criteria for evaluating the packet transfer streams and statistical criteria for evaluating the packet transfer
that mimic TCP over the complete path. The temporal structure of the that mimic TCP over the complete path. The temporal structure of the
test stream (bursts, etc) mimic TCP or other transport protocol test stream (e.g., bursts) mimics TCP or other transport protocols
carrying bulk data over a long path. However they are constructed to carrying bulk data over a long path. However, they are constructed
be independent of the details of the subpath under test, end systems to be independent of the details of the subpath under test, end
or applications. Likewise the success criteria evaluates the packet systems, or applications. Likewise, the success criteria evaluates
transfer statistics of the subpath against criteria determined by the packet transfer statistics of the subpath against criteria
protocol performance models applied to the Target Transport determined by protocol performance models applied to the Target
Performance of the complete path. The success criteria also does not Transport Performance of the complete path. The success criteria
depend on the details of the subpath, end systems or application. also does not depend on the details of the subpath, end systems, or
applications.
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the This document is not an Internet Standards Track specification; it is
provisions of BCP 78 and BCP 79. published for examination, experimental implementation, and
evaluation.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months This document defines an Experimental Protocol for the Internet
and may be updated, replaced, or obsoleted by other documents at any community. This document is a product of the Internet Engineering
time. It is inappropriate to use Internet-Drafts as reference Task Force (IETF). It represents the consensus of the IETF
material or to cite them other than as "work in progress." community. It has received public review and has been approved for
publication by the Internet Engineering Steering Group (IESG). Not
all documents approved by the IESG are candidates for any level of
Internet Standard; see Section 2 of RFC 7841.
This Internet-Draft will expire on March 19, 2018. Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
https://www.rfc-editor.org/info/rfc8337.
Copyright Notice Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the Copyright (c) 2018 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of (https://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License. described in the Simplified BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 1. Introduction ....................................................4
1.1. Version Control . . . . . . . . . . . . . . . . . . . . . 5 2. Overview ........................................................5
2. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3. Terminology .....................................................8
3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 10 3.1. General Terminology ........................................8
4. Background . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2. Terminology about Paths ...................................10
4.1. TCP properties . . . . . . . . . . . . . . . . . . . . . 18 3.3. Properties ................................................11
4.2. Diagnostic Approach . . . . . . . . . . . . . . . . . . . 20 3.4. Basic Parameters ..........................................12
4.3. New requirements relative to RFC 2330 . . . . . . . . . . 21 3.5. Ancillary Parameters ......................................13
5. Common Models and Parameters . . . . . . . . . . . . . . . . 22 3.6. Temporal Patterns for Test Streams ........................14
5.1. Target End-to-end parameters . . . . . . . . . . . . . . 22 3.7. Tests .....................................................15
5.2. Common Model Calculations . . . . . . . . . . . . . . . . 23 4. Background .....................................................16
5.3. Parameter Derating . . . . . . . . . . . . . . . . . . . 24 4.1. TCP Properties ............................................18
5.4. Test Preconditions . . . . . . . . . . . . . . . . . . . 24 4.2. Diagnostic Approach .......................................20
6. Generating test streams . . . . . . . . . . . . . . . . . . . 25 4.3. New Requirements Relative to RFC 2330 .....................21
6.1. Mimicking slowstart . . . . . . . . . . . . . . . . . . . 26 5. Common Models and Parameters ...................................22
6.2. Constant window pseudo CBR . . . . . . . . . . . . . . . 27 5.1. Target End-to-End Parameters ..............................22
6.3. Scanned window pseudo CBR . . . . . . . . . . . . . . . . 28 5.2. Common Model Calculations .................................22
6.4. Concurrent or channelized testing . . . . . . . . . . . . 29 5.3. Parameter Derating ........................................23
7. Interpreting the Results . . . . . . . . . . . . . . . . . . 30 5.4. Test Preconditions ........................................24
7.1. Test outcomes . . . . . . . . . . . . . . . . . . . . . . 30 6. Generating Test Streams ........................................24
7.2. Statistical criteria for estimating run_length . . . . . 31 6.1. Mimicking Slowstart .......................................25
7.3. Reordering Tolerance . . . . . . . . . . . . . . . . . . 34 6.2. Constant Window Pseudo CBR ................................27
8. IP Diagnostic Tests . . . . . . . . . . . . . . . . . . . . . 34 6.3. Scanned Window Pseudo CBR .................................28
8.1. Basic Data Rate and Packet Transfer Tests . . . . . . . . 35 6.4. Concurrent or Channelized Testing .........................28
8.1.1. Delivery Statistics at Paced Full Data Rate . . . . . 35 7. Interpreting the Results .......................................29
8.1.2. Delivery Statistics at Full Data Windowed Rate . . . 35 7.1. Test Outcomes .............................................29
8.1.3. Background Packet Transfer Statistics Tests . . . . . 35 7.2. Statistical Criteria for Estimating run_length ............31
8.2. Standing Queue Tests . . . . . . . . . . . . . . . . . . 36 7.3. Reordering Tolerance ......................................33
8.2.1. Congestion Avoidance . . . . . . . . . . . . . . . . 37 8. IP Diagnostic Tests ............................................34
8.2.2. Bufferbloat . . . . . . . . . . . . . . . . . . . . . 37 8.1. Basic Data Rate and Packet Transfer Tests .................34
8.2.3. Non excessive loss . . . . . . . . . . . . . . . . . 38 8.1.1. Delivery Statistics at Paced Full Data Rate ........35
8.2.4. Duplex Self Interference . . . . . . . . . . . . . . 38 8.1.2. Delivery Statistics at Full Data Windowed Rate .....35
8.3. Slowstart tests . . . . . . . . . . . . . . . . . . . . . 39 8.1.3. Background Packet Transfer Statistics Tests ........35
8.3.1. Full Window slowstart test . . . . . . . . . . . . . 39 8.2. Standing Queue Tests ......................................36
8.3.2. Slowstart AQM test . . . . . . . . . . . . . . . . . 39 8.2.1. Congestion Avoidance ...............................37
8.4. Sender Rate Burst tests . . . . . . . . . . . . . . . . . 40 8.2.2. Bufferbloat ........................................37
8.5. Combined and Implicit Tests . . . . . . . . . . . . . . . 41 8.2.3. Non-excessive Loss .................................38
8.5.1. Sustained Bursts Test . . . . . . . . . . . . . . . . 41 8.2.4. Duplex Self-Interference ...........................38
8.5.2. Passive Measurements . . . . . . . . . . . . . . . . 42 8.3. Slowstart Tests ...........................................39
9. An Example . . . . . . . . . . . . . . . . . . . . . . . . . 43 8.3.1. Full Window Slowstart Test .........................39
9.1. Observations about applicability . . . . . . . . . . . . 44 8.3.2. Slowstart AQM Test .................................39
10. Validation . . . . . . . . . . . . . . . . . . . . . . . . . 44 8.4. Sender Rate Burst Tests ...................................40
11. Security Considerations . . . . . . . . . . . . . . . . . . . 46 8.5. Combined and Implicit Tests ...............................41
12. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 46 8.5.1. Sustained Full-Rate Bursts Test ....................41
13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 47 8.5.2. Passive Measurements ...............................42
14. Informative References . . . . . . . . . . . . . . . . . . . 47
Appendix A. Model Derivations . . . . . . . . . . . . . . . . . 51 9. Example ........................................................43
A.1. Queueless Reno . . . . . . . . . . . . . . . . . . . . . 51 9.1. Observations about Applicability ..........................44
Appendix B. The effects of ACK scheduling . . . . . . . . . . . 52 10. Validation ....................................................45
Appendix C. Version Control . . . . . . . . . . . . . . . . . . 53 11. Security Considerations .......................................46
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 53 12. IANA Considerations ...........................................47
13. Informative References ........................................47
Appendix A. Model Derivations ....................................52
A.1. Queueless Reno ............................................52
Appendix B. The Effects of ACK Scheduling ........................53
Acknowledgments ...................................................55
Authors' Addresses ................................................55
1. Introduction 1. Introduction
Model Based Metrics (MBM) rely on peer-reviewed mathematical models Model-Based Metrics (MBM) rely on peer-reviewed mathematical models
to specify a Targeted Suite of IP Diagnostic tests, designed to to specify a Targeted IP Diagnostic Suite (TIDS), a set of IP
assess whether common transport protocols can be expected to meet a diagnostic tests designed to assess whether common transport
predetermined Target Transport Performance over an Internet path. protocols can be expected to meet a predetermined Target Transport
This note describes the modeling framework to derive the test Performance over an Internet path. This document describes the
parameters for assessing an Internet path's ability to support a modeling framework to derive the test parameters for assessing an
predetermined Bulk Transport Capacity. Internet path's ability to support a predetermined Bulk Transport
Capacity.
Each test in the Targeted IP Diagnostic Suite (TIDS) measures some Each test in TIDS measures some aspect of IP packet transfer needed
aspect of IP packet transfer needed to meet the Target Transport to meet the Target Transport Performance. For Bulk Transport
Performance. For Bulk Transport Capacity the TIDS includes IP Capacity, the TIDS includes IP diagnostic tests to verify that there
diagnostic tests to verify that there is: sufficient IP capacity is sufficient IP capacity (data rate), sufficient queue space at
(data rate); sufficient queue space at bottlenecks to absorb and bottlenecks to absorb and deliver typical transport bursts, low
deliver typical transport bursts; and that the background packet loss enough background packet loss ratio to not interfere with congestion
ratio is low enough not to interfere with congestion control; and control, and other properties described below. Unlike typical IP
other properties described below. Unlike typical IPPM metrics which Performance Metrics (IPPM) that yield measures of network properties,
yield measures of network properties, Model Based Metrics nominally Model-Based Metrics nominally yield pass/fail evaluations of the
yield pass/fail evaluations of the ability of standard transport ability of standard transport protocols to meet the specific
protocols to meet the specific performance objective over some performance objective over some network path.
network path.
In most cases, the IP diagnostic tests can be implemented by In most cases, the IP diagnostic tests can be implemented by
combining existing IPPM metrics with additional controls for combining existing IPPM metrics with additional controls for
generating test streams having a specified temporal structure (bursts generating test streams having a specified temporal structure (bursts
or standing queues caused by constant bit rate streams, etc.) and or standing queues caused by constant bit rate streams, etc.) and
statistical criteria for evaluating packet transfer. The temporal statistical criteria for evaluating packet transfer. The temporal
structure of the test streams mimic transport protocol behavior over structure of the test streams mimics transport protocol behavior over
the complete path; the statistical criteria models the transport the complete path; the statistical criteria models the transport
protocol's response to less than ideal IP packet transfer. In protocol's response to less-than-ideal IP packet transfer. In
control theory terms, the tests are "open loop". Note that running a control theory terms, the tests are "open loop". Note that running a
test requires the coordinated activity of sending and receiving test requires the coordinated activity of sending and receiving
measurement points. measurement points.
This note addresses Bulk Transport Capacity. It describes an This document addresses Bulk Transport Capacity. It describes an
alternative to the approach presented in "A Framework for Defining alternative to the approach presented in "A Framework for Defining
Empirical Bulk Transfer Capacity Metrics" [RFC3148]. Other Model Empirical Bulk Transfer Capacity Metrics" [RFC3148]. Other Model-
Based Metrics may cover other applications and transports, such as Based Metrics may cover other applications and transports, such as
VoIP over UDP and RTP, and new transport protocols. Voice over IP (VoIP) over UDP, RTP, and new transport protocols.
This note assumes a traditional Reno TCP style self clocked, window This document assumes a traditional Reno TCP-style, self-clocked,
controlled transport protocol that uses packet loss and ECN CE marks window-controlled transport protocol that uses packet loss and
for congestion feedback. There are currently some experimental Explicit Congestion Notification (ECN) Congestion Experienced (CE)
marks for congestion feedback. There are currently some experimental
protocols and congestion control algorithms that are rate based or protocols and congestion control algorithms that are rate based or
otherwise fall outside of these assumptions. In the future these new otherwise fall outside of these assumptions. In the future, these
protocols and algorithms may call for revised models. new protocols and algorithms may call for revised models.
The MBM approach, mapping Target Transport Performance to a Targeted The MBM approach, i.e., mapping Target Transport Performance to a
IP Diagnostic Suite (TIDS) of IP tests, solves some intrinsic Targeted IP Diagnostic Suite (TIDS) of IP tests, solves some
problems with using TCP or other throughput maximizing protocols for intrinsic problems with using TCP or other throughput-maximizing
measurement. In particular all throughput maximizing protocols (and protocols for measurement. In particular, all throughput-maximizing
TCP congestion control in particular) cause some level of congestion protocols (especially TCP congestion control) cause some level of
in order to detect when they have reached the available capacity congestion in order to detect when they have reached the available
limitation of the network. This self inflicted congestion obscures capacity limitation of the network. This self-inflicted congestion
the network properties of interest and introduces non-linear dynamic obscures the network properties of interest and introduces non-linear
equilibrium behaviors that make any resulting measurements useless as dynamic equilibrium behaviors that make any resulting measurements
metrics because they have no predictive value for conditions or paths useless as metrics because they have no predictive value for
different than that of the measurement itself. In order to prevent conditions or paths different from that of the measurement itself.
these effects it is necessary to avoid the effects of TCP congestion In order to prevent these effects, it is necessary to avoid the
control in the measurement method. These issues are discussed at effects of TCP congestion control in the measurement method. These
length in Section 4. Readers whom are unfamiliar with basic issues are discussed at length in Section 4. Readers who are
properties of TCP and TCP-like congestion control may find it easier unfamiliar with basic properties of TCP and TCP-like congestion
to start at Section 4 or Section 4.1. control may find it easier to start at Section 4 or 4.1.
A Targeted IP Diagnostic Suite does not have such difficulties. IP A Targeted IP Diagnostic Suite does not have such difficulties. IP
diagnostics can be constructed such that they make strong statistical diagnostics can be constructed such that they make strong statistical
statements about path properties that are independent of the statements about path properties that are independent of measurement
measurement details, such as vantage and choice of measurement details, such as vantage and choice of measurement points.
points.
1.1. Version Control
RFC Editor: Please remove this entire subsection prior to
publication.
REF Editor: The reference to draft-ietf-tcpm-rack is to attribute an
idea. This document should not block waiting for the completion of
that one.
Please send comments about this draft to ippm@ietf.org. See
http://goo.gl/02tkD for more information including: interim drafts,
an up to date todo list and information on contributing.
Formatted: Fri Sep 15 15:07:50 PDT 2017
Changes since -11 draft:
o (From IESG review comments.)
o Ben Campbell: Shorten the Abstract.
o Mirja Kuhlewind: Reduced redundancy. (See message)
o MK: Mention open loop in the introduction.
o MK: Spelled out ECN and reference RFC3168.
o MK: Added a paragraph to the introduction about assuming a
traditional self clocked, window controlled transport protocol.
o MK: Added language about initial window to the list at about
bursts at the end of section 4.1.
o MK: Network power is defined in the terminology section.
o MK: The introduction mention coordinated activity of both
endpoints.
o MK: The security section restates that some of the tests are not
intended for frequent monitoring tests as the high load can impact
other traffic negatively.
o MK: Restored "Informative References" section name.
o And a few minor nits.
Changes since -10 draft:
o A few more nits from various sources.
o (From IETF LC review comments.)
o David Mandelberg: design metrics to prevent DDOS.
o From Robert Sparks:
* Remove all legacy 2119 language.
* Fixed Xr notation inconsistency.
* Adjusted abstract: tests are only partially specified.
* Avoid rather than suppress the effects of congestion control
* Removed the unnecessary, excessively abstract and unclear
thought about IP vs TCP measurements.
* Changed "thwarted" to "not fulfilled".
* Qualified language about burst models.
* Replaced "infinitesimal" with other language.
* Added citations for the reordering strawman.
* Pointed out that pseudo CBR tests depend on self clock.
* Fixed some run on sentences.
o Update language to reflect RFC7567, AQM recommendations.
o Suggestion from Merry Mou (MIT)
Changes since -09 draft:
o Five last minute editing nits.
Changes since -08 draft:
o Language, spelling and usage nits.
o Expanded the abstract describe the models.
o Remove superfluous standards like language
o Remove superfluous "future technology" language.
o Interconnects -> network interconnections.
o Added more labels to Figure 1.
o Defined Bulk Transport.
o Clarified "implied bottleneck IP capacity"
o Clarified the history of the BTC metrics.
o Clarified stochastic vs non-stochastic test traffic generation.
o Reworked Fig 2 and 6.1 "Mimicking slowstart"
o Described the unsynchronized parallel stream failure case.
o Discussed how to measure devices that use virtual queues.
o Changed section 8.5.2 (Streaming Media) to be Passive
Measurements.
Changes since -07 draft:
o Sharpened the use of "statistical criteria"
o Sharpened the definition of test_window, and removed related
redundant text in several places
o Clarified "equilibrium" as "dynamic equilibrium, similar to
processes observed in chemistry"
o Properly explained "Heisenberg" as "observer effect"
o Added the observation from RFC 6576 that HW and SW congestion
control implementations do not generally give the same results.
o Noted that IP and application metrics differ as to how overhead is
handled. MBM is explicit about how it handles overhead.
o Clarified the language and added a new reference about the
problems caused by token bucket policers.
o Added an subsection in the example that comments on some of issues
that need to be mentioned in a future usage or applicability doc.
o Updated ippm-2680-bis to RFC7680
o Many terminology, punctuation and spelling nits.
Changes since -06 draft:
o More language nits:
* "Targeted IP Diagnostic Suite (TIDS)" replaces "Targeted
Diagnostic Suite (TDS)".
* "implied bottleneck IP capacity" replaces "implied bottleneck
IP rate".
* Updated to ECN CE Marks.
* Added "specified temporal structure"
* "test stream" replaces "test traffic"
* "packet transfer" replaces "packet delivery"
* Reworked discussion of slowstart, bursts and pacing.
* RFC 7567 replaces RFC 2309.
Changes since -05 draft:
o Wordsmithing on sections overhauled in -05 draft.
o Reorganized the document:
* Relocated subsection "Preconditions".
* Relocated subsection "New Requirements relative to RFC 2330".
o Addressed nits and not so nits by Ruediger Geib. (Thanks!)
o Substantially tightened the entire definitions section.
o Many terminology changes, to better conform to other docs :
* IP rate and IP capacity (following RFC 5136) replaces various
forms of link data rate.
* subpath replaces link.
* target_window_size replaces target_pipe_size.
* implied bottleneck IP rate replaces effective bottleneck link
rate.
* Packet delivery statistics replaces delivery statistics.
Changes since -04 draft:
o The introduction was heavily overhauled: split into a separate
introduction and overview.
o The new shorter introduction:
* Is a problem statement;
* This document provides a framework;
* That it replaces TCP measurement by IP tests;
* That the results are pass/fail.
o Added a diagram of the framework to the overview
o and introduces all of the elements of the framework.
o Renumbered sections, reducing the depth of some section numbers.
o Updated definitions to better agree with other documents:
* Reordered section 2
* Bulk [data] performance -> Bulk Transport Capacity, everywhere
including the title.
* loss rate and loss probability -> packet loss ratio
* end-to-end path -> complete path
* [end-to-end][target] performance -> Target Transport
Performance
* load test -> capacity test
2. Overview 2. Overview
This document describes a modeling framework for deriving a Targeted This document describes a modeling framework for deriving a Targeted
IP Diagnostic Suite from a predetermined Target Transport IP Diagnostic Suite from a predetermined Target Transport
Performance. It is not a complete specification, and relies on other Performance. It is not a complete specification and relies on other
standards documents to define important details such as packet Type-P standards documents to define important details such as packet type-P
selection, sampling techniques, vantage selection, etc. We imagine selection, sampling techniques, vantage selection, etc. Fully
Fully Specified - Targeted IP Diagnostic Suites (FS-TIDS), that Specified Targeted IP Diagnostic Suites (FSTIDSs) define all of these
define all of these details. We use Targeted IP Diagnostic Suite details. A Targeted IP Diagnostic Suite (TIDS) refers to the subset
(TIDS) to refer to the subset of such a specification that is in of such a specification that is in scope for this document. This
scope for this document. This terminology is defined in Section 3. terminology is further defined in Section 3.
Section 4 describes some key aspects of TCP behavior and what they Section 4 describes some key aspects of TCP behavior and what they
imply about the requirements for IP packet transfer. Most of the IP imply about the requirements for IP packet transfer. Most of the IP
diagnostic tests needed to confirm that the path meets these diagnostic tests needed to confirm that the path meets these
properties can be built on existing IPPM metrics, with the addition properties can be built on existing IPPM metrics, with the addition
of statistical criteria for evaluating packet transfer and in a few of statistical criteria for evaluating packet transfer and, in a few
cases, new mechanisms to implement the required temporal structure. cases, new mechanisms to implement the required temporal structure.
(One group of tests, the standing queue tests described in (One group of tests, the standing queue tests described in
Section 8.2, don't correspond to existing IPPM metrics, but suitable Section 8.2, don't correspond to existing IPPM metrics, but suitable
new IPPM metrics can be patterned after the existing definitions.) new IPPM metrics can be patterned after the existing definitions.)
Figure 1 shows the MBM modeling and measurement framework. The Figure 1 shows the MBM modeling and measurement framework. The
Target Transport Performance, at the top of the figure, is determined Target Transport Performance at the top of the figure is determined
by the needs of the user or application, outside the scope of this by the needs of the user or application, which are outside the scope
document. For Bulk Transport Capacity, the main performance of this document. For Bulk Transport Capacity, the main performance
parameter of interest is the Target Data Rate. However, since TCP's parameter of interest is the Target Data Rate. However, since TCP's
ability to compensate for less than ideal network conditions is ability to compensate for less-than-ideal network conditions is
fundamentally affected by the Round Trip Time (RTT) and the Maximum fundamentally affected by the Round-Trip Time (RTT) and the Maximum
Transmission Unit (MTU) of the complete path, these parameters must Transmission Unit (MTU) of the complete path, these parameters must
also be specified in advance based on knowledge about the intended also be specified in advance based on knowledge about the intended
application setting. They may reflect a specific application over a application setting. They may reflect a specific application over a
real path through the Internet or an idealized application and real path through the Internet or an idealized application and
hypothetical path representing a typical user community. Section 5 hypothetical path representing a typical user community. Section 5
describes the common parameters and models derived from the Target describes the common parameters and models derived from the Target
Transport Performance. Transport Performance.
Target Transport Performance Target Transport Performance
(Target Data Rate, Target RTT and Target MTU) (Target Data Rate, Target RTT, and Target MTU)
| |
________V_________ ________V_________
| mathematical | | mathematical |
| models | | models |
| | | |
------------------ ------------------
Traffic parameters | | Statistical criteria Traffic parameters | | Statistical criteria
| | | |
_______V____________V____Targeted_______ _______V____________V____Targeted IP____
| | * * * | Diagnostic Suite | | | * * * | Diagnostic Suite |
_____|_______V____________V________________ | _____|_______V____________V________________ |
__|____________V____________V______________ | | __|____________V____________V______________ | |
| IP diagnostic tests | | | | IP diagnostic tests | | |
| | | | | | | | | | | |
| _____________V__ __V____________ | | | | _____________V__ __V____________ | | |
| | traffic | | Delivery | | | | | | traffic | | Delivery | | | |
| | pattern | | Evaluation | | | | | | pattern | | Evaluation | | | |
| | generation | | | | | | | | generation | | | | | |
| -------v-------- ------^-------- | | | | -------v-------- ------^-------- | | |
| | v test stream via ^ | | |-- | | v test stream via ^ | | |--
| | -->======================>-- | | | | | -->======================>-- | | |
| | subpath under test | |- | | subpath under test | |-
----V----------------------------------V--- | ----V----------------------------------V--- |
| | | | | | | | | | | |
V V V V V V V V V V V V
fail/inconclusive pass/fail/inconclusive fail/inconclusive pass/fail/inconclusive
(traffic generation status) (test result) (traffic generation status) (test result)
Overall Modeling Framework
Figure 1 Figure 1: Overall Modeling Framework
Mathematical TCP models are used to determine Traffic parameters and Mathematical TCP models are used to determine traffic parameters and
subsequently to design traffic patterns that mimic TCP or other subsequently to design traffic patterns that mimic TCP (which has
transport protocol delivering bulk data and operating at the Target burst characteristics at multiple time scales) or other transport
Data Rate, MTU and RTT over a full range of conditions, including protocols delivering bulk data and operating at the Target Data Rate,
flows that are bursty at multiple time scales. The traffic patterns MTU, and RTT over a full range of conditions. Using the techniques
are generated based on the three Target parameters of complete path described in Section 6, the traffic patterns are generated based on
and independent of the properties of individual subpaths using the the three Target parameters of the complete path (Target Data Rate,
techniques described in Section 6. As much as possible the test Target RTT, and Target MTU), independent of the properties of
streams are generated deterministically (precomputed) to minimize the individual subpaths. As much as possible, the test streams are
extent to which test methodology, measurement points, measurement generated deterministically (precomputed) to minimize the extent to
vantage or path partitioning affect the details of the measurement which test methodology, measurement points, measurement vantage, or
traffic. path partitioning affect the details of the measurement traffic.
Section 7 describes packet transfer statistics and methods to test Section 7 describes packet transfer statistics and methods to test
them against the statistical criteria provided by the mathematical against the statistical criteria provided by the mathematical models.
models. Since the statistical criteria typically apply to the Since the statistical criteria typically apply to the complete path
complete path (a composition of subpaths) [RFC6049], in situ testing (a composition of subpaths) [RFC6049], in situ testing requires that
requires that the end-to-end statistical criteria be apportioned as the end-to-end statistical criteria be apportioned as separate
separate criteria for each subpath. Subpaths that are expected to be criteria for each subpath. Subpaths that are expected to be
bottlenecks would then be permitted to contribute a larger fraction bottlenecks would then be permitted to contribute a larger fraction
of the end-to-end packet loss budget. In compensation, subpaths that of the end-to-end packet loss budget. In compensation, subpaths that
are not expected to exhibit bottlenecks must be constrained to are not expected to exhibit bottlenecks must be constrained to
contribute less packet loss. Thus the statistical criteria for each contribute less packet loss. Thus, the statistical criteria for each
subpath in each test of a TIDS is an apportioned share of the end-to- subpath in each test of a TIDS is an apportioned share of the end-to-
end statistical criteria for the complete path which was determined end statistical criteria for the complete path that was determined by
by the mathematical model. the mathematical model.
Section 8 describes the suite of individual tests needed to verify Section 8 describes the suite of individual tests needed to verify
all of required IP delivery properties. A subpath passes if and only all of the required IP delivery properties. A subpath passes if and
if all of the individual IP diagnostic tests pass. Any subpath that only if all of the individual IP diagnostic tests pass. Any subpath
fails any test indicates that some users are likely to fail to attain that fails any test indicates that some users are likely to fail to
their Target Transport Performance under some conditions. In attain their Target Transport Performance under some conditions. In
addition to passing or failing, a test can be deemed to be addition to passing or failing, a test can be deemed inconclusive for
inconclusive for a number of reasons including: the precomputed a number of reasons, including the following: the precomputed traffic
traffic pattern was not accurately generated; the measurement results pattern was not accurately generated, the measurement results were
were not statistically significant; and others such as failing to not statistically significant, the test failed to meet some required
meet some required test preconditions. If all tests pass but some test preconditions, etc. If all tests pass but some are
are inconclusive, then the entire suite is deemed to be inconclusive. inconclusive, then the entire suite is deemed to be inconclusive.
In Section 9 we present an example TIDS that might be representative In Section 9, we present an example TIDS that might be representative
of High Definition (HD) video, and illustrate how Model Based Metrics of High Definition (HD) video and illustrate how Model-Based Metrics
can be used to address difficult measurement situations, such as can be used to address difficult measurement situations, such as
confirming that inter-carrier exchanges have sufficient performance confirming that inter-carrier exchanges have sufficient performance
and capacity to deliver HD video between ISPs. and capacity to deliver HD video between ISPs.
Since there is some uncertainty in the modeling process, Section 10 Since there is some uncertainty in the modeling process, Section 10
describes a validation procedure to diagnose and minimize false describes a validation procedure to diagnose and minimize false
positive and false negative results. positive and false negative results.
3. Terminology 3. Terminology
Terms containing underscores (rather than spaces) appear in equations Terms containing underscores (rather than spaces) appear in equations
and typically have algorithmic definitions. and typically have algorithmic definitions.
General Terminology: 3.1. General Terminology
Target: A general term for any parameter specified by or derived Target: A general term for any parameter specified by or derived
from the user's application or transport performance requirements. from the user's application or transport performance requirements.
Target Transport Performance: Application or transport performance Target Transport Performance: Application or transport performance
target values for the complete path. For Bulk Transport Capacity target values for the complete path. For Bulk Transport Capacity
defined in this note the Target Transport Performance includes the defined in this document, the Target Transport Performance
Target Data Rate, Target RTT and Target MTU as described below. includes the Target Data Rate, Target RTT, and Target MTU as
described below.
Target Data Rate: The specified application data rate required for Target Data Rate: The specified application data rate required for
an application's proper operation. Conventional Bulk Transport an application's proper operation. Conventional Bulk Transport
Capacity (BTC) metrics are focused on the Target Data Rate, Capacity (BTC) metrics are focused on the Target Data Rate;
however these metrics had little or no predictive value because however, these metrics have little or no predictive value because
they do not consider the effects of the other two parameters of they do not consider the effects of the other two parameters of
the Target Transport Performance, the RTT and MTU of the complete the Target Transport Performance -- the RTT and MTU of the
paths. complete paths.
Target RTT (Round Trip Time): The specified baseline (minimum) RTT
Target RTT (Round-Trip Time): The specified baseline (minimum) RTT
of the longest complete path over which the user expects to be of the longest complete path over which the user expects to be
able to meet the target performance. TCP and other transport able to meet the target performance. TCP and other transport
protocol's ability to compensate for path problems is generally protocol's ability to compensate for path problems is generally
proportional to the number of round trips per second. The Target proportional to the number of round trips per second. The Target
RTT determines both key parameters of the traffic patterns (e.g. RTT determines both key parameters of the traffic patterns (e.g.,
burst sizes) and the thresholds on acceptable IP packet transfer burst sizes) and the thresholds on acceptable IP packet transfer
statistics. The Target RTT must be specified considering statistics. The Target RTT must be specified considering
appropriate packets sizes: MTU sized packets on the forward path, appropriate packets sizes: MTU-sized packets on the forward path
ACK sized packets (typically header_overhead) on the return path. and ACK-sized packets (typically, header_overhead) on the return
Note that Target RTT is specified and not measured, MBM path. Note that Target RTT is specified and not measured; MBM
measurements derived for a given target_RTT will be applicable to measurements derived for a given target_RTT will be applicable to
any path with a smaller RTTs. any path with a smaller RTT.
Target MTU (Maximum Transmission Unit): The specified maximum MTU Target MTU (Maximum Transmission Unit): The specified maximum MTU
supported by the complete path the over which the application supported by the complete path over which the application expects
expects to meet the target performance. In this document assume a to meet the target performance. In this document, we assume a
1500 Byte MTU unless otherwise specified. If some subpath has a 1500-byte MTU unless otherwise specified. If a subpath has a
smaller MTU, then it becomes the Target MTU for the complete path, smaller MTU, then it becomes the Target MTU for the complete path,
and all model calculations and subpath tests must use the same and all model calculations and subpath tests must use the same
smaller MTU. smaller MTU.
Targeted IP Diagnostic Suite (TIDS): A set of IP diagnostic tests Targeted IP Diagnostic Suite (TIDS): A set of IP diagnostic tests
designed to determine if an otherwise ideal complete path designed to determine if an otherwise ideal complete path
containing the subpath under test can sustain flows at a specific containing the subpath under test can sustain flows at a specific
target_data_rate using target_MTU sized packets when the RTT of target_data_rate using packets with a size of target_MTU when the
the complete path is target_RTT. RTT of the complete path is target_RTT.
Fully Specified Targeted IP Diagnostic Suite (FS-TIDS): A TIDS
together with additional specification such as measurement packet Fully Specified Targeted IP Diagnostic Suite (FSTIDS): A TIDS
type ("type-p" [RFC2330]), etc. which are out of scope for this together with additional specifications such as measurement packet
document, but need to be drawn from other standards documents. type ("type-p" [RFC2330]) that are out of scope for this document
Bulk Transport Capacity: Bulk Transport Capacity Metrics evaluate an and need to be drawn from other standards documents.
Internet path's ability to carry bulk data, such as large files,
streaming (non-real time) video, and under some conditions, web Bulk Transport Capacity (BTC): Bulk Transport Capacity metrics
images and other content. Prior efforts to define BTC metrics evaluate an Internet path's ability to carry bulk data, such as
have been based on [RFC3148], which predates our understanding of large files, streaming (non-real-time) video, and, under some
TCP and the requirements described in Section 4. In general "Bulk conditions, web images and other content. Prior efforts to define
Transport" indicates that performance is determined by the BTC metrics have been based on [RFC3148], which predates our
interplay between the network, cross traffic and congestion understanding of TCP and the requirements described in Section 4.
control in the transport protocol. It excludes situations where In general, "Bulk Transport" indicates that performance is
performance is dominated by the RTT alone (e.g. transactions) or determined by the interplay between the network, cross traffic,
bottlenecks elsewhere, such as in the application itself. and congestion control in the transport protocol. It excludes
situations where performance is dominated by the RTT alone (e.g.,
transactions) or bottlenecks elsewhere, such as in the application
itself.
IP diagnostic tests: Measurements or diagnostics to determine if IP diagnostic tests: Measurements or diagnostics to determine if
packet transfer statistics meet some precomputed target. packet transfer statistics meet some precomputed target.
traffic patterns: The temporal patterns or burstiness of traffic traffic patterns: The temporal patterns or burstiness of traffic
generated by applications over transport protocols such as TCP. generated by applications over transport protocols such as TCP.
There are several mechanisms that cause bursts at various time There are several mechanisms that cause bursts at various
scales as described in Section 4.1. Our goal here is to mimic the timescales as described in Section 4.1. Our goal here is to mimic
range of common patterns (burst sizes and rates, etc), without the range of common patterns (burst sizes, rates, etc.), without
tying our applicability to specific applications, implementations tying our applicability to specific applications, implementations,
or technologies, which are sure to become stale. or technologies, which are sure to become stale.
Explicit Congestion Notification (ECN): See [RFC3168]. Explicit Congestion Notification (ECN): See [RFC3168].
packet transfer statistics: Raw, detailed or summary statistics
packet transfer statistics: Raw, detailed, or summary statistics
about packet transfer properties of the IP layer including packet about packet transfer properties of the IP layer including packet
losses, ECN Congestion Experienced (CE) marks, reordering, or any losses, ECN Congestion Experienced (CE) marks, reordering, or any
other properties that may be germane to transport performance. other properties that may be germane to transport performance.
packet loss ratio: As defined in [RFC7680]. packet loss ratio: As defined in [RFC7680].
apportioned: To divide and allocate, for example budgeting packet
apportioned: To divide and allocate, for example, budgeting packet
loss across multiple subpaths such that the losses will accumulate loss across multiple subpaths such that the losses will accumulate
to less than a specified end-to-end loss ratio. Apportioning to less than a specified end-to-end loss ratio. Apportioning
metrics is essentially the inverse of the process described in metrics is essentially the inverse of the process described in
[RFC5835]. [RFC5835].
open loop: A control theory term used to describe a class of open loop: A control theory term used to describe a class of
techniques where systems that naturally exhibit circular techniques where systems that naturally exhibit circular
dependencies can be analyzed by suppressing some of the dependencies can be analyzed by suppressing some of the
dependencies, such that the resulting dependency graph is acyclic. dependencies, such that the resulting dependency graph is acyclic.
Terminology about paths, etc. See [RFC2330] and [RFC7398] for 3.2. Terminology about Paths
existing terms and definitions.
See [RFC2330] and [RFC7398] for existing terms and definitions.
data sender: Host sending data and receiving ACKs. data sender: Host sending data and receiving ACKs.
data receiver: Host receiving data and sending ACKs. data receiver: Host receiving data and sending ACKs.
complete path: The end-to-end path from the data sender to the data complete path: The end-to-end path from the data sender to the data
receiver. receiver.
subpath: A portion of the complete path. Note that there is no subpath: A portion of the complete path. Note that there is no
requirement that subpaths be non-overlapping. A subpath can be a requirement that subpaths be non-overlapping. A subpath can be as
small as a single device, link or interface. small as a single device, link, or interface.
measurement point: Measurement points as described in [RFC7398]. measurement point: Measurement points as described in [RFC7398].
test path: A path between two measurement points that includes a test path: A path between two measurement points that includes a
subpath of the complete path under test. If the measurement subpath of the complete path under test. If the measurement
points are off path, the test path may include "test leads" points are off path, the test path may include "test leads"
between the measurement points and the subpath. between the measurement points and the subpath.
dominant bottleneck: The bottleneck that generally determines most dominant bottleneck: The bottleneck that generally determines most
of packet transfer statistics for the entire path. It typically packet transfer statistics for the entire path. It typically
determines a flow's self clock timing, packet loss and ECN determines a flow's self-clock timing, packet loss, and ECN CE
Congestion Experienced (CE) marking rate, with other potential marking rate, with other potential bottlenecks having less effect
bottlenecks having less effect on the packet transfer statistics. on the packet transfer statistics. See Section 4.1 on TCP
See Section 4.1 on TCP properties. properties.
front path: The subpath from the data sender to the dominant front path: The subpath from the data sender to the dominant
bottleneck. bottleneck.
back path: The subpath from the dominant bottleneck to the receiver. back path: The subpath from the dominant bottleneck to the receiver.
return path: The path taken by the ACKs from the data receiver to return path: The path taken by the ACKs from the data receiver to
the data sender. the data sender.
cross traffic: Other, potentially interfering, traffic competing for cross traffic: Other, potentially interfering, traffic competing for
network resources (bandwidth and/or queue capacity). network resources (such as bandwidth and/or queue capacity).
Properties determined by the complete path and application. These 3.3. Properties
are described in more detail in Section 5.1.
The following properties are determined by the complete path and
application. These are described in more detail in Section 5.1.
Application Data Rate: General term for the data rate as seen by the Application Data Rate: General term for the data rate as seen by the
application above the transport layer in bytes per second. This application above the transport layer in bytes per second. This
is the payload data rate, and explicitly excludes transport and is the payload data rate and explicitly excludes transport-level
lower level headers (TCP/IP or other protocols), retransmissions and lower-level headers (TCP/IP or other protocols),
and other overhead that is not part to the total quantity of data retransmissions, and other overhead that is not part of the total
delivered to the application. quantity of data delivered to the application.
IP rate: The actual number of IP-layer bytes delivered through a IP rate: The actual number of IP-layer bytes delivered through a
subpath, per unit time, including TCP and IP headers, retransmits subpath, per unit time, including TCP and IP headers, retransmits,
and other TCP/IP overhead. Follows from IP-type-P Link Usage and other TCP/IP overhead. This is the same as IP-type-P Link
[RFC5136]. Usage in [RFC5136].
IP capacity: The maximum number of IP-layer bytes that can be IP capacity: The maximum number of IP-layer bytes that can be
transmitted through a subpath, per unit time, including TCP and IP transmitted through a subpath, per unit time, including TCP and IP
headers, retransmits and other TCP/IP overhead. Follows from IP- headers, retransmits, and other TCP/IP overhead. This is the same
type-P Link Capacity [RFC5136]. as IP-type-P Link Capacity in [RFC5136].
bottleneck IP capacity: The IP capacity of the dominant bottleneck bottleneck IP capacity: The IP capacity of the dominant bottleneck
in the forward path. All throughput maximizing protocols estimate in the forward path. All throughput-maximizing protocols estimate
this capacity by observing the IP rate delivered through the this capacity by observing the IP rate delivered through the
bottleneck. Most protocols derive their self clocks from the bottleneck. Most protocols derive their self-clocks from the
timing of this data. See Section 4.1 and Appendix B for more timing of this data. See Section 4.1 and Appendix B for more
details. details.
implied bottleneck IP capacity: This is the bottleneck IP capacity
implied by the ACKs returning from the receiver. It is determined implied bottleneck IP capacity: The bottleneck IP capacity implied
by looking at how much application data the ACK stream at the by the ACKs returning from the receiver. It is determined by
sender reports delivered to the data receiver per unit time at looking at how much application data the ACK stream at the sender
various time scales. If the return path is thinning, batching or reports as delivered to the data receiver per unit time at various
otherwise altering the ACK timing the implied bottleneck IP timescales. If the return path is thinning, batching, or
capacity over short time scales might be substantially larger than otherwise altering the ACK timing, the implied bottleneck IP
capacity over short timescales might be substantially larger than
the bottleneck IP capacity averaged over a full RTT. Since TCP the bottleneck IP capacity averaged over a full RTT. Since TCP
derives its clock from the data delivered through the bottleneck, derives its clock from the data delivered through the bottleneck,
the front path must have sufficient buffering to absorb any data the front path must have sufficient buffering to absorb any data
bursts at the dimensions (size and IP rate) implied by the ACK bursts at the dimensions (size and IP rate) implied by the ACK
stream, which are potentially doubled during slowstart. If the stream, which are potentially doubled during slowstart. If the
return path is not altering the ACK stream, then the implied return path is not altering the ACK stream, then the implied
bottleneck IP capacity will be the same as the bottleneck IP bottleneck IP capacity will be the same as the bottleneck IP
capacity. See Section 4.1 and Appendix B for more details. capacity. See Section 4.1 and Appendix B for more details.
sender interface rate: The IP rate which corresponds to the IP sender interface rate: The IP rate that corresponds to the IP
capacity of the data sender's interface. Due to sender efficiency capacity of the data sender's interface. Due to sender efficiency
algorithms including technologies such as TCP segmentation offload algorithms, including technologies such as TCP segmentation
(TSO), nearly all modern servers deliver data in bursts at full offload (TSO), nearly all modern servers deliver data in bursts at
interface link rate. Today 1 or 10 Gb/s are typical. full interface link rate. Today, 1 or 10 Gb/s are typical.
Header_overhead: The IP and TCP header sizes, which are the portion
header_overhead: The IP and TCP header sizes, which are the portion
of each MTU not available for carrying application payload. of each MTU not available for carrying application payload.
Without loss of generality this is assumed to be the size for Without loss of generality, this is assumed to be the size for
returning acknowledgments (ACKs). For TCP, the Maximum Segment returning acknowledgments (ACKs). For TCP, the Maximum Segment
Size (MSS) is the Target MTU minus the header_overhead. Size (MSS) is the Target MTU minus the header_overhead.
Basic parameters common to models and subpath tests are defined here 3.4. Basic Parameters
are described in more detail in Section 5.2. Note that these are
mixed between application transport performance (excludes headers) Basic parameters common to models and subpath tests are defined here.
and IP performance (which include TCP headers and retransmissions as Formulas for target_window_size and target_run_length appear in
part of the IP payload). Section 5.2. Note that these are mixed between application transport
performance (excludes headers) and IP performance (includes TCP
headers and retransmissions as part of the IP payload).
Network power: The observed data rate divided by the observed RTT. Network power: The observed data rate divided by the observed RTT.
Network power indicates how effectively a transport protocol is Network power indicates how effectively a transport protocol is
filling a network. filling a network.
Window [size]: The total quantity of data carried by packets in-
flight plus the data represented by ACKs circulating in the Window [size]: The total quantity of data carried by packets
in-flight plus the data represented by ACKs circulating in the
network is referred to as the window. See Section 4.1. Sometimes network is referred to as the window. See Section 4.1. Sometimes
used with other qualifiers (congestion window, cwnd or receiver used with other qualifiers (congestion window (cwnd) or receiver
window) to indicate which mechanism is controlling the window. window) to indicate which mechanism is controlling the window.
pipe size: A general term for number of packets needed in flight
(the window size) to exactly fill some network path or subpath. pipe size: A general term for the number of packets needed in flight
It corresponds to the window size which maximizes network power. (the window size) to exactly fill a network path or subpath. It
Often used with additional qualifiers to specify which path, or corresponds to the window size, which maximizes network power. It
is often used with additional qualifiers to specify which path,
under what conditions, etc. under what conditions, etc.
target_window_size: The average number of packets in flight (the target_window_size: The average number of packets in flight (the
window size) needed to meet the Target Data Rate, for the window size) needed to meet the Target Data Rate for the specified
specified Target RTT, and MTU. It implies the scale of the bursts Target RTT and Target MTU. It implies the scale of the bursts
that the network might experience. that the network might experience.
run length: A general term for the observed, measured, or specified run length: A general term for the observed, measured, or specified
number of packets that are (expected to be) delivered between number of packets that are (expected to be) delivered between
losses or ECN Congestion Experienced (CE) marks. Nominally one losses or ECN CE marks. Nominally, it is one over the sum of the
over the sum of the loss and ECN CE marking probabilities, if loss and ECN CE marking probabilities, if they are independently
there are independently and identically distributed. and identically distributed.
target_run_length: The target_run_length is an estimate of the target_run_length: The target_run_length is an estimate of the
minimum number of non-congestion marked packets needed between minimum number of non-congestion marked packets needed between
losses or ECN Congestion Experienced (CE) marks necessary to losses or ECN CE marks necessary to attain the target_data_rate
attain the target_data_rate over a path with the specified over a path with the specified target_RTT and target_MTU, as
target_RTT and target_MTU, as computed by a mathematical model of computed by a mathematical model of TCP congestion control. A
TCP congestion control. A reference calculation is shown in reference calculation is shown in Section 5.2 and alternatives in
Section 5.2 and alternatives in Appendix A Appendix A.
reference target_run_length: target_run_length computed precisely by reference target_run_length: target_run_length computed precisely by
the method in Section 5.2. This is likely to be slightly more the method in Section 5.2. This is likely to be slightly more
conservative than required by modern TCP implementations. conservative than required by modern TCP implementations.
Ancillary parameters used for some tests: 3.5. Ancillary Parameters
derating: Under some conditions the standard models are too The following ancillary parameters are used for some tests:
derating: Under some conditions, the standard models are too
conservative. The modeling framework permits some latitude in conservative. The modeling framework permits some latitude in
relaxing or "derating" some test parameters as described in relaxing or "derating" some test parameters, as described in
Section 5.3 in exchange for a more stringent TIDS validation Section 5.3, in exchange for a more stringent TIDS validation
procedures, described in Section 10. Models can be derated by procedures, described in Section 10. Models can be derated by
including a multiplicative derating factor to make tests less including a multiplicative derating factor to make tests less
stringent. stringent.
subpath_IP_capacity: The IP capacity of a specific subpath. subpath_IP_capacity: The IP capacity of a specific subpath.
test path: A subpath of a complete path under test. test path: A subpath of a complete path under test.
test_path_RTT: The RTT observed between two measurement points using test_path_RTT: The RTT observed between two measurement points using
packet sizes that are consistent with the transport protocol. packet sizes that are consistent with the transport protocol.
This is generally MTU sized packets of the forward path, This is generally MTU-sized packets of the forward path and
header_overhead sized packets on the return path. packets with a size of header_overhead on the return path.
test_path_pipe: The pipe size of a test path. Nominally the
test_path_pipe: The pipe size of a test path. Nominally, it is the
test_path_RTT times the test path IP_capacity. test_path_RTT times the test path IP_capacity.
test_window: The smallest window sufficient to meet or exceed the test_window: The smallest window sufficient to meet or exceed the
target_rate when operating with a pure self clock over a test target_rate when operating with a pure self-clock over a test
path. The test_window is typically given by path. The test_window is typically calculated as follows (but see
ceiling(target_data_rate*test_path_RTT/(target_MTU- the discussion in Appendix B about the effects of channel
header_overhead)) but see the discussion in Appendix B about the scheduling on RTT):
effects of channel scheduling on RTT. On some test paths the
test_window may need to be adjusted slightly to compensate for the ceiling(target_data_rate * test_path_RTT / (target_MTU -
RTT being inflated by the devices that schedule packets. header_overhead))
On some test paths, the test_window may need to be adjusted
slightly to compensate for the RTT being inflated by the devices
that schedule packets.
3.6. Temporal Patterns for Test Streams
The terminology below is used to define temporal patterns for test The terminology below is used to define temporal patterns for test
stream. These patterns are designed to mimic TCP behavior, as streams. These patterns are designed to mimic TCP behavior, as
described in Section 4.1. described in Section 4.1.
packet headway: Time interval between packets, specified from the packet headway: Time interval between packets, specified from the
start of one to the start of the next. e.g. If packets are sent start of one to the start of the next. For example, if packets
with a 1 mS headway, there will be exactly 1000 packets per are sent with a 1 ms headway, there will be exactly 1000 packets
second. per second.
burst headway: Time interval between bursts, specified from the burst headway: Time interval between bursts, specified from the
start of the first packet one burst to the start of the first start of the first packet of one burst to the start of the first
packet of the next burst. e.g. If 4 packet bursts are sent with a packet of the next burst. For example, if 4 packet bursts are
1 mS burst headway, there will be exactly 4000 packets per second. sent with a 1 ms burst headway, there will be exactly 4000 packets
paced single packets: Send individual packets at the specified rate per second.
paced single packets: Individual packets sent at the specified rate
or packet headway. or packet headway.
paced bursts: Send bursts on a timer. Specify any 3 of: average
data rate, packet size, burst size (number of packets) and burst paced bursts: Bursts on a timer. Specify any 3 of the following:
headway (burst start to start). By default the bursts are assumed average data rate, packet size, burst size (number of packets),
to occur at full sender interface rate, such that the packet and burst headway (burst start to start). By default, the bursts
headway within each burst is the minimum supported by the sender's are assumed to occur at full sender interface rate, such that the
interface. Under some conditions it is useful to explicitly packet headway within each burst is the minimum supported by the
specify the packet headway within each burst. sender's interface. Under some conditions, it is useful to
slowstart rate: Mimic TCP slowstart by sending 4 packet paced bursts explicitly specify the packet headway within each burst.
at an average data rate equal to twice the implied bottleneck IP
capacity (but not more than the sender interface rate). This is a slowstart rate: Paced bursts of four packets each at an average data
two level burst pattern described in more detail in Section 6.1. rate equal to twice the implied bottleneck IP capacity (but not
If the implied bottleneck IP capacity is more than half of the more than the sender interface rate). This mimics TCP slowstart.
sender interface rate, slowstart rate becomes sender interface This is a two-level burst pattern described in more detail in
rate. Section 6.1. If the implied bottleneck IP capacity is more than
slowstart burst: Mimic one round of TCP slowstart by sending a half of the sender interface rate, the slowstart rate becomes the
specified number of packets packets in a two level burst pattern sender interface rate.
that resembles slowstart.
repeated slowstart bursts: Repeat Slowstart bursts once per slowstart burst: A specified number of packets in a two-level burst
target_RTT. For TCP each burst would be twice as large as the pattern that resembles slowstart. This mimics one round of TCP
slowstart.
repeated slowstart bursts: Slowstart bursts repeated once per
target_RTT. For TCP, each burst would be twice as large as the
prior burst, and the sequence would end at the first ECN CE mark prior burst, and the sequence would end at the first ECN CE mark
or lost packet. For measurement, all slowstart bursts would be or lost packet. For measurement, all slowstart bursts would be
the same size (nominally target_window_size but other sizes might the same size (nominally, target_window_size but other sizes might
be specified), and the ECN CE marks and lost packets are counted. be specified), and the ECN CE marks and lost packets are counted.
The tests described in this note can be grouped according to their 3.7. Tests
applicability.
The tests described in this document can be grouped according to
their applicability.
Capacity tests: Capacity tests determine if a network subpath has Capacity tests: Capacity tests determine if a network subpath has
sufficient capacity to deliver the Target Transport Performance. sufficient capacity to deliver the Target Transport Performance.
As long as the test stream is within the proper envelope for the As long as the test stream is within the proper envelope for the
Target Transport Performance, the average packet losses or ECN Target Transport Performance, the average packet losses or ECN CE
Congestion Experienced (CE) marks must be below the statistical marks must be below the statistical criteria computed by the
criteria computed by the model. As such, capacity tests reflect model. As such, capacity tests reflect parameters that can
parameters that can transition from passing to failing as a transition from passing to failing as a consequence of cross
consequence of cross traffic, additional presented load or the traffic, additional presented load, or the actions of other
actions of other network users. By definition, capacity tests network users. By definition, capacity tests also consume
also consume significant network resources (data capacity and/or significant network resources (data capacity and/or queue buffer
queue buffer space), and the test schedules must be balanced by space), and the test schedules must be balanced by their cost.
their cost.
Monitoring tests: Monitoring tests are designed to capture the most Monitoring tests: Monitoring tests are designed to capture the most
important aspects of a capacity test, but without presenting important aspects of a capacity test without presenting excessive
excessive ongoing load themselves. As such they may miss some ongoing load themselves. As such, they may miss some details of
details of the network's performance, but can serve as a useful the network's performance but can serve as a useful reduced-cost
reduced-cost proxy for a capacity test, for example to support proxy for a capacity test, for example, to support continuous
continuous production network monitoring. production network monitoring.
Engineering tests: Engineering tests evaluate how network algorithms Engineering tests: Engineering tests evaluate how network algorithms
(such as AQM and channel allocation) interact with TCP-style self (such as Active Queue Management (AQM) and channel allocation)
clocked protocols and adaptive congestion control based on packet interact with TCP-style self-clocked protocols and adaptive
loss and ECN Congestion Experienced (CE) marks. These tests are congestion control based on packet loss and ECN CE marks. These
likely to have complicated interactions with cross traffic and tests are likely to have complicated interactions with cross
under some conditions can be inversely sensitive to load. For traffic and, under some conditions, can be inversely sensitive to
example a test to verify that an AQM algorithm causes ECN CE marks load. For example, a test to verify that an AQM algorithm causes
or packet drops early enough to limit queue occupancy may ECN CE marks or packet drops early enough to limit queue occupancy
experience a false pass result in the presence of cross traffic. may experience a false pass result in the presence of cross
It is important that engineering tests be performed under a wide traffic. It is important that engineering tests be performed
range of conditions, including both in situ and bench testing, and under a wide range of conditions, including both in situ and bench
over a wide variety of load conditions. Ongoing monitoring is testing, and over a wide variety of load conditions. Ongoing
less likely to be useful for engineering tests, although sparse in monitoring is less likely to be useful for engineering tests,
situ testing might be appropriate. although sparse in situ testing might be appropriate.
4. Background 4. Background
At the time the "Framework for IP Performance Metrics" [RFC2330] was When "Framework for IP Performance Metrics" [RFC2330] was published
published (1998), sound Bulk Transport Capacity (BTC) measurement was in 1998, sound Bulk Transport Capacity (BTC) measurement was known to
known to be well beyond our capabilities. Even when Framework for be well beyond our capabilities. Even when "A Framework for Defining
Empirical BTC Metrics [RFC3148] was published, we knew that we didn't Empirical Bulk Transfer Capacity Metrics" [RFC3148] was published, we
really understand the problem. Now, by hindsight we understand why knew that we didn't really understand the problem. Now, in
assessing BTC is such a hard problem: hindsight, we understand why assessing BTC is such a difficult
problem:
o TCP is a control system with circular dependencies - everything o TCP is a control system with circular dependencies -- everything
affects performance, including components that are explicitly not affects performance, including components that are explicitly not
part of the test (for example, the host processing power is not part of the test (for example, the host processing power is not
in-scope of path performance tests). in-scope of path performance tests).
o Congestion control is a dynamic equilibrium process, similar to o Congestion control is a dynamic equilibrium process, similar to
processes observed in chemistry and other fields. The network and processes observed in chemistry and other fields. The network and
transport protocols find an operating point which balances between transport protocols find an operating point that balances opposing
opposing forces: the transport protocol pushing harder (raising forces: the transport protocol pushing harder (raising the data
the data rate and/or window) while the network pushes back rate and/or window) while the network pushes back (raising packet
(raising packet loss ratio, RTT and/or ECN CE marks). By design loss ratio, RTT, and/or ECN CE marks). By design, TCP congestion
TCP congestion control keeps raising the data rate until the control keeps raising the data rate until the network gives some
network gives some indication that its capacity has been exceeded indication that its capacity has been exceeded by dropping packets
by dropping packets or adding ECN CE marks. If a TCP sender or adding ECN CE marks. If a TCP sender accurately fills a path
accurately fills a path to its IP capacity, (e.g. the bottleneck to its IP capacity (e.g., the bottleneck is 100% utilized), then
is 100% utilized), then packet losses and ECN CE marks are mostly packet losses and ECN CE marks are mostly determined by the TCP
determined by the TCP sender and how aggressively it seeks sender and how aggressively it seeks additional capacity; they are
additional capacity, and not the network itself, since the network not determined by the network itself, because the network must
must send exactly the signals that TCP needs to set its rate. send exactly the signals that TCP needs to set its rate.
o TCP's ability to compensate for network impairments (such as loss, o TCP's ability to compensate for network impairments (such as loss,
delay and delay variation, outside of those caused by TCP itself) delay, and delay variation, outside of those caused by TCP itself)
is directly proportional to the number of send-ACK round trip is directly proportional to the number of send-ACK round-trip
exchanges per second (i.e. inversely proportional to the RTT). As exchanges per second (i.e., inversely proportional to the RTT).
a consequence an impaired subpath may pass a short RTT local test As a consequence, an impaired subpath may pass a short RTT local
even though it fails when the subpath is extended by an test even though it fails when the subpath is extended by an
effectively perfect network to some larger RTT. effectively perfect network to some larger RTT.
o TCP has an extreme form of the Observer Effect (colloquially know
as the Heisenberg effect). Measurement and cross traffic interact o TCP has an extreme form of the Observer Effect (colloquially known
in unknown and ill defined ways. The situation is actually worse as the "Heisenberg Effect"). Measurement and cross traffic
than the traditional physics problem where you can at least interact in unknown and ill-defined ways. The situation is
estimate bounds on the relative momentum of the measurement and actually worse than the traditional physics problem where you can
measured particles. For network measurement you can not in at least estimate bounds on the relative momentum of the
general determine even the order of magnitude of the effect. It measurement and measured particles. In general, for network
is possible to construct measurement scenarios where the measurement, you cannot determine even the order of magnitude of
measurement traffic starves real user traffic, yielding an overly the effect. It is possible to construct measurement scenarios
inflated measurement. The inverse is also possible: the user where the measurement traffic starves real user traffic, yielding
traffic can fill the network, such that the measurement traffic an overly inflated measurement. The inverse is also possible: the
detects only minimal available capacity. You can not in general user traffic can fill the network, such that the measurement
determine which scenario might be in effect, so you can not gauge traffic detects only minimal available capacity. In general, you
the relative magnitude of the uncertainty introduced by cannot determine which scenario might be in effect, so you cannot
gauge the relative magnitude of the uncertainty introduced by
interactions with other network traffic. interactions with other network traffic.
o As a consequence of the properties listed above it is difficult,
if not impossible, for two independent implementations (HW or SW) o As a consequence of the properties listed above, it is difficult,
of TCP congestion control to produce equivalent performance if not impossible, for two independent implementations (hardware
results [RFC6576] under the same network conditions, or software) of TCP congestion control to produce equivalent
performance results [RFC6576] under the same network conditions.
These properties are a consequence of the dynamic equilibrium These properties are a consequence of the dynamic equilibrium
behavior intrinsic to how all throughput maximizing protocols behavior intrinsic to how all throughput-maximizing protocols
interact with the Internet. These protocols rely on control systems interact with the Internet. These protocols rely on control systems
based on estimated network metrics to regulate the quantity of data based on estimated network metrics to regulate the quantity of data
to send into the network. The packet sending characteristics in turn to send into the network. The packet-sending characteristics in turn
alter the network properties estimated by the control system metrics, alter the network properties estimated by the control system metrics,
such that there are circular dependencies between every transmission such that there are circular dependencies between every transmission
characteristic and every estimated metric. Since some of these characteristic and every estimated metric. Since some of these
dependencies are nonlinear, the entire system is nonlinear, and any dependencies are nonlinear, the entire system is nonlinear, and any
change anywhere causes a difficult to predict response in network change anywhere causes a difficult-to-predict response in network
metrics. As a consequence Bulk Transport Capacity metrics have not metrics. As a consequence, Bulk Transport Capacity metrics have not
fulfilled the analytic framework envisioned in [RFC2330] fulfilled the analytic framework envisioned in [RFC2330].
Model Based Metrics overcome these problems by making the measurement Model-Based Metrics overcome these problems by making the measurement
system open loop: the packet transfer statistics (akin to the network system open loop: the packet transfer statistics (akin to the network
estimators) do not affect the traffic or traffic patterns (bursts), estimators) do not affect the traffic or traffic patterns (bursts),
which are computed on the basis of the Target Transport Performance. which are computed on the basis of the Target Transport Performance.
A path or subpath meeting the Target Transfer Performance A path or subpath meeting the Target Transfer Performance
requirements would exhibit packet transfer statistics and estimated requirements would exhibit packet transfer statistics and estimated
metrics that would not cause the control system to slow the traffic metrics that would not cause the control system to slow the traffic
below the Target Data Rate. below the Target Data Rate.
4.1. TCP properties 4.1. TCP Properties
TCP and other self clocked protocols (e.g. SCTP) carry the vast TCP and other self-clocked protocols (e.g., the Stream Control
majority of all Internet data. Their dominant bulk data transport Transmission Protocol (SCTP)) carry the vast majority of all Internet
behavior is to have an approximately fixed quantity of data and data. Their dominant bulk data transport behavior is to have an
acknowledgments (ACKs) circulating in the network. The data receiver approximately fixed quantity of data and acknowledgments (ACKs)
reports arriving data by returning ACKs to the data sender, the data circulating in the network. The data receiver reports arriving data
sender typically responds by sending approximately the same quantity by returning ACKs to the data sender, and the data sender typically
of data back into the network. The total quantity of data plus the responds by sending approximately the same quantity of data back into
data represented by ACKs circulating in the network is referred to as the network. The total quantity of data plus the data represented by
the window. The mandatory congestion control algorithms ACKs circulating in the network is referred to as the "window". The
incrementally adjust the window by sending slightly more or less data mandatory congestion control algorithms incrementally adjust the
in response to each ACK. The fundamentally important property of window by sending slightly more or less data in response to each ACK.
this system is that it is self clocked: The data transmissions are a The fundamentally important property of this system is that it is
reflection of the ACKs that were delivered by the network, the ACKs self-clocked: the data transmissions are a reflection of the ACKs
are a reflection of the data arriving from the network. that were delivered by the network, and the ACKs are a reflection of
the data arriving from the network.
A number of protocol features cause bursts of data, even in idealized A number of protocol features cause bursts of data, even in idealized
networks that can be modeled as simple queuing systems. networks that can be modeled as simple queuing systems.
During slowstart the IP rate is doubled on each RTT by sending twice During slowstart, the IP rate is doubled on each RTT by sending twice
as much data as was delivered to the receiver during the prior RTT. as much data as was delivered to the receiver during the prior RTT.
Each returning ACK causes the sender to transmit twice the data the Each returning ACK causes the sender to transmit twice the data the
ACK reported arriving at the receiver. For slowstart to be able to ACK reported arriving at the receiver. For slowstart to be able to
fill the pipe, the network must be able to tolerate slowstart bursts fill the pipe, the network must be able to tolerate slowstart bursts
up to the full pipe size inflated by the anticipated window reduction up to the full pipe size inflated by the anticipated window reduction
on the first loss or ECN CE mark. For example, with classic Reno on the first loss or ECN CE mark. For example, with classic Reno
congestion control, an optimal slowstart has to end with a burst that congestion control, an optimal slowstart has to end with a burst that
is twice the bottleneck rate for one RTT in duration. This burst is twice the bottleneck rate for one RTT in duration. This burst
causes a queue which is equal to the pipe size (i.e. the window is causes a queue that is equal to the pipe size (i.e., the window is
twice the pipe size) so when the window is halved in response to the twice the pipe size), so when the window is halved in response to the
first packet loss, the new window will be the pipe size. first packet loss, the new window will be the pipe size.
Note that if the bottleneck IP rate is less that half of the capacity Note that if the bottleneck IP rate is less than half of the capacity
of the front path (which is almost always the case), the slowstart of the front path (which is almost always the case), the slowstart
bursts will not by themselves cause significant queues anywhere else bursts will not by themselves cause significant queues anywhere else
along the front path; they primarily exercise the queue at the along the front path; they primarily exercise the queue at the
dominant bottleneck. dominant bottleneck.
Several common efficiency algorithms also cause bursts. The self Several common efficiency algorithms also cause bursts. The self-
clock is typically applied to groups of packets: the receiver's clock is typically applied to groups of packets: the receiver's
delayed ACK algorithm generally sends only one ACK per two data delayed ACK algorithm generally sends only one ACK per two data
segments. Furthermore the modern senders use TCP segmentation segments. Furthermore, modern senders use TCP segmentation offload
offload (TSO) to reduce CPU overhead. The sender's software stack (TSO) to reduce CPU overhead. The sender's software stack builds
builds super sized TCP segments that the TSO hardware splits into MTU super-sized TCP segments that the TSO hardware splits into MTU-sized
sized segments on the wire. The net effect of TSO, delayed ACK and segments on the wire. The net effect of TSO, delayed ACK, and other
other efficiency algorithms is to send bursts of segments at full efficiency algorithms is to send bursts of segments at full sender
sender interface rate. interface rate.
Note that these efficiency algorithms are almost always in effect, Note that these efficiency algorithms are almost always in effect,
including during slowstart, such that slowstart typically has a two including during slowstart, such that slowstart typically has a two-
level burst structure. Section 6.1 describes slowstart in more level burst structure. Section 6.1 describes slowstart in more
detail. detail.
Additional sources of bursts include TCP's initial window [RFC6928], Additional sources of bursts include TCP's initial window [RFC6928],
application pauses, channel allocation mechanisms and network devices application pauses, channel allocation mechanisms, and network
that schedule ACKs. Appendix B describes these last two items. If devices that schedule ACKs. Appendix B describes these last two
the application pauses (stops reading or writing data) for some items. If the application pauses (e.g., stops reading or writing
fraction of an RTT, many TCP implementations catch up to their data) for some fraction of an RTT, many TCP implementations catch up
earlier window size by sending a burst of data at the full sender to their earlier window size by sending a burst of data at the full
interface rate. To fill a network with a realistic application, the sender interface rate. To fill a network with a realistic
network has to be able to tolerate sender interface rate bursts large application, the network has to be able to tolerate sender interface
enough to restore the prior window following application pauses. rate bursts large enough to restore the prior window following
application pauses.
Although the sender interface rate bursts are typically smaller than Although the sender interface rate bursts are typically smaller than
the last burst of a slowstart, they are at a higher IP rate so they the last burst of a slowstart, they are at a higher IP rate so they
potentially exercise queues at arbitrary points along the front path potentially exercise queues at arbitrary points along the front path
from the data sender up to and including the queue at the dominant from the data sender up to and including the queue at the dominant
bottleneck. It is known that these bursts can hurt network bottleneck. It is known that these bursts can hurt network
performance, especially in conjunction with other queue pressure, performance, especially in conjunction with other queue pressure;
however we are not aware of any models for how frequent sender rate however, we are not aware of any models for estimating the impact or
bursts the network should be able to tolerate at various burst sizes. prescribing limits on the size or frequency of sender rate bursts.
In conclusion, to verify that a path can meet a Target Transport In conclusion, to verify that a path can meet a Target Transport
Performance, it is necessary to independently confirm that the path Performance, it is necessary to independently confirm that the path
can tolerate bursts at the scales that can be caused by the above can tolerate bursts at the scales that can be caused by the above
mechanisms. Three cases are believed to be sufficient: mechanisms. Three cases are believed to be sufficient:
o Two level slowstart bursts sufficient to get connections started o Two-level slowstart bursts sufficient to get connections started
properly. properly.
o Ubiquitous sender interface rate bursts caused by efficiency o Ubiquitous sender interface rate bursts caused by efficiency
algorithms. We assume 4 packet bursts to be the most common case, algorithms. We assume four packet bursts to be the most common
since it matches the effects of delayed ACK during slowstart. case, since it matches the effects of delayed ACK during
These bursts should be assumed not to significantly affect packet slowstart. These bursts should be assumed not to significantly
transfer statistics. affect packet transfer statistics.
o Infrequent sender interface rate bursts that are the maximum of o Infrequent sender interface rate bursts that are the maximum of
the full target_window_size and the initial window size (10 the full target_window_size and the initial window size (10
segments in [RFC6928]). The Target_run_length may be derated for segments in [RFC6928]). The target_run_length may be derated for
these large fast bursts. these large fast bursts.
If a subpath can meet the required packet loss ratio for bursts at If a subpath can meet the required packet loss ratio for bursts at
all of these scales then it has sufficient buffering at all potential all of these scales, then it has sufficient buffering at all
bottlenecks to tolerate any of the bursts that are likely introduced potential bottlenecks to tolerate any of the bursts that are likely
by TCP or other transport protocols. introduced by TCP or other transport protocols.
4.2. Diagnostic Approach 4.2. Diagnostic Approach
A complete path of a given RTT and MTU, which are equal to or smaller A complete path is expected to be able to attain a specified Bulk
than the Target RTT and equal to or larger than the Target MTU Transport Capacity if the path's RTT is equal to or smaller than the
respectively, is expected to be able to attain a specified Bulk Target RTT, the path's MTU is equal to or larger than the Target MTU,
Transport Capacity when all of the following conditions are met: and all of the following conditions are met:
1. The IP capacity is above the Target Data Rate by sufficient 1. The IP capacity is above the Target Data Rate by a sufficient
margin to cover all TCP/IP overheads. This can be confirmed by margin to cover all TCP/IP overheads. This can be confirmed by
the tests described in Section 8.1 or any number of IP capacity the tests described in Section 8.1 or any number of IP capacity
tests adapted to implement MBM. tests adapted to implement MBM.
2. The observed packet transfer statistics are better than required 2. The observed packet transfer statistics are better than required
by a suitable TCP performance model (e.g. fewer packet losses or by a suitable TCP performance model (e.g., fewer packet losses or
ECN CE marks). See Section 8.1 or any number of low or fixed ECN CE marks). See Section 8.1 or any number of low- or fixed-
rate packet loss tests outside of MBM. rate packet loss tests outside of MBM.
3. There is sufficient buffering at the dominant bottleneck to 3. There is sufficient buffering at the dominant bottleneck to
absorb a slowstart bursts large enough to get the flow out of absorb a slowstart burst large enough to get the flow out of
slowstart at a suitable window size. See Section 8.3. slowstart at a suitable window size. See Section 8.3.
4. There is sufficient buffering in the front path to absorb and 4. There is sufficient buffering in the front path to absorb and
smooth sender interface rate bursts at all scales that are likely smooth sender interface rate bursts at all scales that are likely
to be generated by the application, any channel arbitration in to be generated by the application, any channel arbitration in
the ACK path or any other mechanisms. See Section 8.4. the ACK path, or any other mechanisms. See Section 8.4.
5. When there is a slowly rising standing queue at the bottleneck
the onset of packet loss has to be at an appropriate point (time 5. When there is a slowly rising standing queue at the bottleneck,
or queue depth) and progressive [RFC7567]. See Section 8.2. then the onset of packet loss has to be at an appropriate point
(in time or in queue depth) and has to be progressive, for
example, by use of Active Queue Management [RFC7567]. See
Section 8.2.
6. When there is a standing queue at a bottleneck for a shared media 6. When there is a standing queue at a bottleneck for a shared media
subpath (e.g. half duplex), there must be a suitable bounds on subpath (e.g., a half-duplex link), there must be a suitable
the interaction between ACKs and data, for example due to the bound on the interaction between ACKs and data, for example, due
channel arbitration mechanism. See Section 8.2.4. to the channel arbitration mechanism. See Section 8.2.4.
Note that conditions 1 through 4 require capacity tests for Note that conditions 1 through 4 require capacity tests for
validation, and thus may need to be monitored on an ongoing basis. validation and thus may need to be monitored on an ongoing basis.
Conditions 5 and 6 require engineering tests, which are best Conditions 5 and 6 require engineering tests, which are best
performed in controlled environments such as a bench test. They performed in controlled environments (e.g., bench tests). They won't
won't generally fail due to load, but may fail in the field due to generally fail due to load but may fail in the field (e.g., due to
configuration errors, etc. and should be spot checked. configuration errors, etc.) and thus should be spot checked.
A tool that can perform many of the tests is available from A tool that can perform many of the tests is available from
[MBMSource]. [MBMSource].
4.3. New requirements relative to RFC 2330 4.3. New Requirements Relative to RFC 2330
Model Based Metrics are designed to fulfill some additional Model-Based Metrics are designed to fulfill some additional
requirements that were not recognized at the time RFC 2330 was requirements that were not recognized at the time RFC 2330 [RFC2330]
written [RFC2330]. These missing requirements may have significantly was published. These missing requirements may have significantly
contributed to policy difficulties in the IP measurement space. Some contributed to policy difficulties in the IP measurement space. Some
additional requirements are: additional requirements are:
o IP metrics must be actionable by the ISP - they have to be o IP metrics must be actionable by the ISP -- they have to be
interpreted in terms of behaviors or properties at the IP or lower interpreted in terms of behaviors or properties at the IP or lower
layers, that an ISP can test, repair and verify. layers that an ISP can test, repair, and verify.
o Metrics should be spatially composable, such that measures of o Metrics should be spatially composable, such that measures of
concatenated paths should be predictable from subpaths. concatenated paths should be predictable from subpaths.
o Metrics must be vantage point invariant over a significant range o Metrics must be vantage point invariant over a significant range
of measurement point choices, including off path measurement of measurement point choices, including off-path measurement
points. The only requirements on MP selection should be that the points. The only requirements for Measurement Point (MP)
RTT between the MPs is below some reasonable bound, and that the selection should be that the RTT between the MPs is below some
effects of the "test leads" connecting MPs to the subpath under reasonable bound and that the effects of the "test leads"
test can be can be calibrated out of the measurements. The latter connecting MPs to the subpath under test can be calibrated out of
might be be accomplished if the test leads are effectively ideal the measurements. The latter might be accomplished if the test
or their properties can be deducted from the measurements between leads are effectively ideal or their properties can be deducted
the MPs. While many of tests require that the test leads have at from the measurements between the MPs. While many tests require
least as much IP capacity as the subpath under test, some do not, that the test leads have at least as much IP capacity as the
for example Background Packet Transfer Tests described in subpath under test, some do not, for example, the Background
Section 8.1.3. Packet Transfer Statistics Tests described in Section 8.1.3.
o Metric measurements should be repeatable by multiple parties with o Metric measurements should be repeatable by multiple parties with
no specialized access to MPs or diagnostic infrastructure. It no specialized access to MPs or diagnostic infrastructure. It
should be possible for different parties to make the same should be possible for different parties to make the same
measurement and observe the same results. In particular it is measurement and observe the same results. In particular, it is
specifically important that both a consumer (or their delegate) important that both a consumer (or the consumer's delegate) and
and ISP be able to perform the same measurement and get the same ISP be able to perform the same measurement and get the same
result. Note that vantage independence is key to meeting this result. Note that vantage independence is key to meeting this
requirement. requirement.
5. Common Models and Parameters 5. Common Models and Parameters
5.1. Target End-to-end parameters 5.1. Target End-to-End Parameters
The target end-to-end parameters are the Target Data Rate, Target RTT The target end-to-end parameters are the Target Data Rate, Target
and Target MTU as defined in Section 3. These parameters are RTT, and Target MTU as defined in Section 3. These parameters are
determined by the needs of the application or the ultimate end user determined by the needs of the application or the ultimate end user
and the complete Internet path over which the application is expected and the complete Internet path over which the application is expected
to operate. The target parameters are in units that make sense to to operate. The target parameters are in units that make sense to
upper layers: payload bytes delivered to the application, above TCP. layers above the TCP layer: payload bytes delivered to the
They exclude overheads associated with TCP and IP headers, application. They exclude overheads associated with TCP and IP
retransmits and other protocols (e.g. DNS). Note that IP-based headers, retransmits and other protocols (e.g., DNS). Note that
network services include TCP headers and retransmissions as part of IP-based network services include TCP headers and retransmissions as
delivered payload, and this difference is recognized in calculations part of delivered payload; this difference (header_overhead) is
below (header_overhead). recognized in calculations below.
Other end-to-end parameters defined in Section 3 include the Other end-to-end parameters defined in Section 3 include the
effective bottleneck data rate, the sender interface data rate and effective bottleneck data rate, the sender interface data rate, and
the TCP and IP header sizes. the TCP and IP header sizes.
The target_data_rate must be smaller than all subpath IP capacities The target_data_rate must be smaller than all subpath IP capacities
by enough headroom to carry the transport protocol overhead, by enough headroom to carry the transport protocol overhead,
explicitly including retransmissions and an allowance for explicitly including retransmissions and an allowance for
fluctuations in TCP's actual data rate. Specifying a fluctuations in TCP's actual data rate. Specifying a
target_data_rate with insufficient headroom is likely to result in target_data_rate with insufficient headroom is likely to result in
brittle measurements having little predictive value. brittle measurements that have little predictive value.
Note that the target parameters can be specified for a hypothetical Note that the target parameters can be specified for a hypothetical
path, for example to construct TIDS designed for bench testing in the path (for example, to construct TIDS designed for bench testing in
absence of a real application; or for a live in situ test of the absence of a real application) or for a live in situ test of
production infrastructure. production infrastructure.
The number of concurrent connections is explicitly not a parameter to The number of concurrent connections is explicitly not a parameter in
this model. If a subpath requires multiple connections in order to this model. If a subpath requires multiple connections in order to
meet the specified performance, that must be stated explicitly and meet the specified performance, that must be stated explicitly, and
the procedure described in Section 6.4 applies. the procedure described in Section 6.4 applies.
5.2. Common Model Calculations 5.2. Common Model Calculations
The Target Transport Performance is used to derive the The Target Transport Performance is used to derive the
target_window_size and the reference target_run_length. target_window_size and the reference target_run_length.
The target_window_size, is the average window size in packets needed The target_window_size is the average window size in packets needed
to meet the target_rate, for the specified target_RTT and target_MTU. to meet the target_rate, for the specified target_RTT and target_MTU.
It is given by: To calculate target_window_size:
target_window_size = ceiling( target_rate * target_RTT / ( target_MTU
- header_overhead ) )
Target_run_length is an estimate of the minimum required number of target_window_size = ceiling(target_rate * target_RTT / (target_MTU -
unmarked packets that must be delivered between losses or ECN header_overhead))
Congestion Experienced (CE) marks, as computed by a mathematical The target_run_length is an estimate of the minimum required number
model of TCP congestion control. The derivation here follows of unmarked packets that must be delivered between losses or ECN CE
[MSMO97], and by design is quite conservative. marks, as computed by a mathematical model of TCP congestion control.
The derivation here is parallel to the derivation in [MSMO97] and, by
design, is quite conservative.
Reference target_run_length is derived as follows: assume the The reference target_run_length is derived as follows. Assume the
subpath_IP_capacity is infinitesimally larger than the subpath_IP_capacity is infinitesimally larger than the
target_data_rate plus the required header_overhead. Then target_data_rate plus the required header_overhead. Then,
target_window_size also predicts the onset of queuing. A larger target_window_size also predicts the onset of queuing. A larger
window will cause a standing queue at the bottleneck. window will cause a standing queue at the bottleneck.
Assume the transport protocol is using standard Reno style Additive Assume the transport protocol is using standard Reno-style Additive
Increase, Multiplicative Decrease (AIMD) congestion control [RFC5681] Increase Multiplicative Decrease (AIMD) congestion control [RFC5681]
(but not Appropriate Byte Counting [RFC3465]) and the receiver is (but not Appropriate Byte Counting [RFC3465]) and the receiver is
using standard delayed ACKs. Reno increases the window by one packet using standard delayed ACKs. Reno increases the window by one packet
every pipe_size worth of ACKs. With delayed ACKs this takes 2 Round every pipe size worth of ACKs. With delayed ACKs, this takes two
Trip Times per increase. To exactly fill the pipe, the spacing of RTTs per increase. To exactly fill the pipe, the spacing of losses
losses must be no closer than when the peak of the AIMD sawtooth must be no closer than when the peak of the AIMD sawtooth reached
reached exactly twice the target_window_size. Otherwise, the exactly twice the target_window_size. Otherwise, the multiplicative
multiplicative window reduction triggered by the loss would cause the window reduction triggered by the loss would cause the network to be
network to be under-filled. Following [MSMO97] the number of packets underfilled. Per [MSMO97] the number of packets between losses must
between losses must be the area under the AIMD sawtooth. They must be the area under the AIMD sawtooth. They must be no more frequent
be no more frequent than every 1 in than every 1 in ((3/2)*target_window_size)*(2*target_window_size)
((3/2)*target_window_size)*(2*target_window_size) packets, which packets, which simplifies to:
simplifies to:
target_run_length = 3*(target_window_size^2) target_run_length = 3*(target_window_size^2)
Note that this calculation is very conservative and is based on a Note that this calculation is very conservative and is based on a
number of assumptions that may not apply. Appendix A discusses these number of assumptions that may not apply. Appendix A discusses these
assumptions and provides some alternative models. If a different assumptions and provides some alternative models. If a different
model is used, a FS-TIDS must document the actual method for model is used, an FSTIDS must document the actual method for
computing target_run_length and ratio between alternate computing target_run_length and the ratio between alternate
target_run_length and the reference target_run_length calculated target_run_length and the reference target_run_length calculated
above, along with a discussion of the rationale for the underlying above, along with a discussion of the rationale for the underlying
assumptions. assumptions.
These two parameters, target_window_size and target_run_length, Most of the individual parameters for the tests in Section 8 are
directly imply most of the individual parameters for the tests in derived from target_window_size and target_run_length.
Section 8.
5.3. Parameter Derating 5.3. Parameter Derating
Since some aspects of the models are very conservative, the MBM Since some aspects of the models are very conservative, the MBM
framework permits some latitude in derating test parameters. Rather framework permits some latitude in derating test parameters. Rather
than trying to formalize more complicated models we permit some test than trying to formalize more complicated models, we permit some test
parameters to be relaxed as long as they meet some additional parameters to be relaxed as long as they meet some additional
procedural constraints: procedural constraints:
o The FS-TIDS must document and justify the actual method used to o The FSTIDS must document and justify the actual method used to
compute the derated metric parameters. compute the derated metric parameters.
o The validation procedures described in Section 10 must be used to o The validation procedures described in Section 10 must be used to
demonstrate the feasibility of meeting the Target Transport demonstrate the feasibility of meeting the Target Transport
Performance with infrastructure that just barely passes the Performance with infrastructure that just barely passes the
derated tests. derated tests.
o The validation process for a FS-TIDS itself must be documented is
o The validation process for an FSTIDS itself must be documented in
such a way that other researchers can duplicate the validation such a way that other researchers can duplicate the validation
experiments. experiments.
Except as noted, all tests below assume no derating. Tests where Except as noted, all tests below assume no derating. Tests for which
there is not currently a well established model for the required there is not currently a well-established model for the required
parameters explicitly include derating as a way to indicate parameters explicitly include derating as a way to indicate
flexibility in the parameters. flexibility in the parameters.
5.4. Test Preconditions 5.4. Test Preconditions
Many tests have preconditions which are required to assure their Many tests have preconditions that are required to assure their
validity. Examples include: the presence or non-presence of cross validity. Examples include the presence or non-presence of cross
traffic on specific subpaths; negotiating ECN; and appropriate traffic on specific subpaths; negotiating ECN; and a test stream
preamble packet stream to testing to put reactive network elements preamble of appropriate length to achieve stable access to network
into the proper states [RFC7312]. If preconditions are not properly resources in the presence of reactive network elements (as defined in
Section 1.1 of [RFC7312]). If preconditions are not properly
satisfied for some reason, the tests should be considered to be satisfied for some reason, the tests should be considered to be
inconclusive. In general it is useful to preserve diagnostic inconclusive. In general, it is useful to preserve diagnostic
information as to why the preconditions were not met, and any test information as to why the preconditions were not met and any test
data that was collected even if it is not useful for the intended data that was collected even if it is not useful for the intended
test. Such diagnostic information and partial test data may be test. Such diagnostic information and partial test data may be
useful for improving the test or test procedures themselves. useful for improving the test or test procedures themselves.
It is important to preserve the record that a test was scheduled, It is important to preserve the record that a test was scheduled;
because otherwise precondition enforcement mechanisms can introduce otherwise, precondition enforcement mechanisms can introduce sampling
sampling bias. For example, canceling tests due to cross traffic on bias. For example, canceling tests due to cross traffic on
subscriber access links might introduce sampling bias in tests of the subscriber access links might introduce sampling bias in tests of the
rest of the network by reducing the number of tests during peak rest of the network by reducing the number of tests during peak
network load. network load.
Test preconditions and failure actions must be specified in a FS- Test preconditions and failure actions must be specified in an
TIDS. FSTIDS.
6. Generating test streams 6. Generating Test Streams
Many important properties of Model Based Metrics, such as vantage Many important properties of Model-Based Metrics, such as vantage
independence, are a consequence of using test streams that have independence, are a consequence of using test streams that have
temporal structures that mimic TCP or other transport protocols temporal structures that mimic TCP or other transport protocols
running over a complete path. As described in Section 4.1, self running over a complete path. As described in Section 4.1, self-
clocked protocols naturally have burst structures related to the RTT clocked protocols naturally have burst structures related to the RTT
and pipe size of the complete path. These bursts naturally get and pipe size of the complete path. These bursts naturally get
larger (contain more packets) as either the Target RTT or Target Data larger (contain more packets) as either the Target RTT or Target Data
Rate get larger, or the Target MTU gets smaller. An implication of Rate get larger or the Target MTU gets smaller. An implication of
these relationships is that test streams generated by running self these relationships is that test streams generated by running self-
clocked protocols over short subpaths may not adequately exercise the clocked protocols over short subpaths may not adequately exercise the
queuing at any bottleneck to determine if the subpath can support the queuing at any bottleneck to determine if the subpath can support the
full Target Transport Performance over the complete path. full Target Transport Performance over the complete path.
Failing to authentically mimic TCP's temporal structure is part of Failing to authentically mimic TCP's temporal structure is part of
the reason why simple performance tools such as iPerf, netperf, nc, the reason why simple performance tools such as iPerf, netperf, nc,
etc have the reputation of yielding false pass results over short etc., have the reputation for yielding false pass results over short
test paths, even when some subpath has a flaw. test paths, even when a subpath has a flaw.
The definitions in Section 3 are sufficient for most test streams. The definitions in Section 3 are sufficient for most test streams.
We describe the slowstart and standing queue test streams in more We describe the slowstart and standing queue test streams in more
detail. detail.
In conventional measurement practice stochastic processes are used to In conventional measurement practice, stochastic processes are used
eliminate many unintended correlations and sample biases. However to eliminate many unintended correlations and sample biases.
MBM tests are designed to explicitly mimic temporal correlations However, MBM tests are designed to explicitly mimic temporal
caused by network or protocol elements themselves. Some portions of correlations caused by network or protocol elements themselves. Some
these systems, such as traffic arrival (test scheduling) are portions of these systems, such as traffic arrival (e.g., test
naturally stochastic. Other behaviors, such as back-to-back packet scheduling), are naturally stochastic. Other behaviors, such as
transmissions, are dominated by implementation specific deterministic back-to-back packet transmissions, are dominated by implementation-
effects. Although these behaviors always contain non-deterministic specific deterministic effects. Although these behaviors always
elements and might be modeled stochastically, these details typically contain non-deterministic elements and might be modeled
do not contribute significantly to the overall system behavior. stochastically, these details typically do not contribute
Furthermore, it is known that real protocols are subject to failures significantly to the overall system behavior. Furthermore, it is
caused by network property estimators suffering from bias due to known that real protocols are subject to failures caused by network
correlation in their own traffic. For example TCP's RTT estimator property estimators suffering from bias due to correlation in their
used to determine the Retransmit Time Out (RTO), can be fooled by own traffic. For example, TCP's RTT estimator used to determine the
periodic cross traffic or start-stop applications. For these reasons Retransmit Timeout (RTO), can be fooled by periodic cross traffic or
many details of the test streams are specified deterministically. start-stop applications. For these reasons, many details of the test
streams are specified deterministically.
It may prove useful to introduce fine grained noise sources into the It may prove useful to introduce fine-grained noise sources into the
models used for generating test streams in an update of Model Based models used for generating test streams in an update of Model-Based
Metrics, but the complexity is not warranted at the time this Metrics, but the complexity is not warranted at the time this
document was written. document was written.
6.1. Mimicking slowstart 6.1. Mimicking Slowstart
TCP slowstart has a two level burst structure as shown in Figure 2. TCP slowstart has a two-level burst structure as shown in Figure 2.
The fine time structure is caused by efficiency algorithms that The fine time structure is caused by efficiency algorithms that
deliberately batch work (CPU, channel allocation, etc) to better deliberately batch work (CPU, channel allocation, etc.) to better
amortize certain network and host overheads. ACKs passing through amortize certain network and host overheads. ACKs passing through
the return path typically cause the sender to transmit small bursts the return path typically cause the sender to transmit small bursts
of data at full sender interface rate. For example TCP Segmentation of data at the full sender interface rate. For example, TCP
Offload (TSO) and Delayed Acknowledgment both contribute to this Segmentation Offload (TSO) and Delayed Acknowledgment both contribute
effect. During slowstart these bursts are at the same headway as the to this effect. During slowstart, these bursts are at the same
returning ACKs, but are typically twice as large (e.g. having twice headway as the returning ACKs but are typically twice as large (e.g.,
as much data) as the ACK reported was delivered to the receiver. Due have twice as much data) as the ACK reported was delivered to the
to variations in delayed ACK and algorithms such as Appropriate Byte receiver. Due to variations in delayed ACK and algorithms such as
Counting [RFC3465], different pairs of senders and receivers produce Appropriate Byte Counting [RFC3465], different pairs of senders and
slightly different burst patterns. Without loss of generality, we receivers produce slightly different burst patterns. Without loss of
assume each ACK causes 4 packet sender interface rate bursts at an generality, we assume each ACK causes four packet sender interface
average headway equal to the ACK headway, and corresponding to rate bursts at an average headway equal to the ACK headway; this
sending at an average rate equal to twice the effective bottleneck IP corresponds to sending at an average rate equal to twice the
rate. Each slowstart burst consists of a series of 4 packet sender effective bottleneck IP rate. Each slowstart burst consists of a
interface rate bursts such that the total number of packets is the series of four packet sender interface rate bursts such that the
current window size (as of the last packet in the burst). total number of packets is the current window size (as of the last
packet in the burst).
The coarse time structure is due to each RTT being a reflection of The coarse time structure is due to each RTT being a reflection of
the prior RTT. For real transport protocols, each slowstart burst is the prior RTT. For real transport protocols, each slowstart burst is
twice as large (twice the window) as the previous burst but is spread twice as large (twice the window) as the previous burst but is spread
out in time by the network bottleneck, such that each successive RTT out in time by the network bottleneck, such that each successive RTT
exhibits the same effective bottleneck IP rate. The slowstart phase exhibits the same effective bottleneck IP rate. The slowstart phase
ends on the first lost packet or ECN mark, which is intended to ends on the first lost packet or ECN mark, which is intended to
happen after successive slowstart bursts merge in time: the next happen after successive slowstart bursts merge in time: the next
burst starts before the bottleneck queue is fully drained and the burst starts before the bottleneck queue is fully drained and the
prior burst is complete. prior burst is complete.
For diagnostic tests described below we preserve the fine time For the diagnostic tests described below, we preserve the fine time
structure but manipulate the coarse structure of the slowstart bursts structure but manipulate the coarse structure of the slowstart bursts
(burst size and headway) to measure the ability of the dominant (burst size and headway) to measure the ability of the dominant
bottleneck to absorb and smooth slowstart bursts. bottleneck to absorb and smooth slowstart bursts.
Note that a stream of repeated slowstart bursts has three different Note that a stream of repeated slowstart bursts has three different
average rates, depending on the averaging time interval. At the average rates, depending on the averaging time interval. At the
finest time scale (a few packet times at the sender interface) the finest timescale (a few packet times at the sender interface), the
peak of the average IP rate is the same as the sender interface rate; peak of the average IP rate is the same as the sender interface rate;
at a medium timescale (a few ACK times at the dominant bottleneck) at a medium timescale (a few ACK times at the dominant bottleneck),
the peak of the average IP rate is twice the implied bottleneck IP the peak of the average IP rate is twice the implied bottleneck IP
capacity; and at time scales longer than the target_RTT and when the capacity; and at timescales longer than the target_RTT and when the
burst size is equal to the target_window_size, the average rate is burst size is equal to the target_window_size, the average rate is
equal to the target_data_rate. This pattern corresponds to repeating equal to the target_data_rate. This pattern corresponds to repeating
the last RTT of TCP slowstart when delayed ACK and sender side byte the last RTT of TCP slowstart when delayed ACK and sender-side byte
counting are present but without the limits specified in Appropriate counting are present but without the limits specified in Appropriate
Byte Counting [RFC3465]. Byte Counting [RFC3465].
time ==> ( - equals one packet) time ==> ( - equals one packet)
Fine time structure of the packet stream: Fine time structure of the packet stream:
---- ---- ---- ---- ---- ---- ---- ---- ---- ----
|<>| sender interface rate bursts (typically 3 or 4 packets) |<>| sender interface rate bursts (typically 3 or 4 packets)
|<===>| burst headway (from the ACK headway) |<===>| burst headway (from the ACK headway)
\____repeating sender______/ \____repeating sender______/
rate bursts rate bursts
Coarse (RTT level) time structure of the packet stream: Coarse (RTT-level) time structure of the packet stream:
---- ---- ---- ---- ---- ---- ---- ... ---- ---- ---- ---- ---- ---- ---- ...
|<========================>| slowstart burst size (from the window) |<========================>| slowstart burst size (from the window)
|<==============================================>| slowstart headway |<==============================================>| slowstart headway
(from the RTT) (from the RTT)
\__________________________/ \_________ ... \__________________________/ \_________ ...
one slowstart burst Repeated slowstart bursts one slowstart burst Repeated slowstart bursts
Multiple levels of Slowstart Bursts Figure 2: Multiple Levels of Slowstart Bursts
Figure 2
6.2. Constant window pseudo CBR 6.2. Constant Window Pseudo CBR
Implement pseudo constant bit rate by running a standard self clocked Pseudo constant bit rate (CBR) is implemented by running a standard
protocol such as TCP with a fixed window size. If that window size self-clocked protocol such as TCP with a fixed window size. If that
is test_window, the data rate will be slightly above the target_rate. window size is test_window, the data rate will be slightly above the
target_rate.
Since the test_window is constrained to be an integer number of Since the test_window is constrained to be an integer number of
packets, for small RTTs or low data rates there may not be packets, for small RTTs or low data rates, there may not be
sufficiently precise control over the data rate. Rounding the sufficiently precise control over the data rate. Rounding the
test_window up (as defined above) is likely to result in data rates test_window up (as defined above) is likely to result in data rates
that are higher than the target rate, but reducing the window by one that are higher than the target rate, but reducing the window by one
packet may result in data rates that are too small. Also cross packet may result in data rates that are too small. Also, cross
traffic potentially raises the RTT, implicitly reducing the rate. traffic potentially raises the RTT, implicitly reducing the rate.
Cross traffic that raises the RTT nearly always makes the test more Cross traffic that raises the RTT nearly always makes the test more
strenuous (more demanding for the network path). strenuous (i.e., more demanding for the network path).
Note that Constant window pseudo CBR (and Scanned window pseudo CBR Note that Constant Window Pseudo CBR (and Scanned Window Pseudo CBR
in the next section) both rely on a self clock which is at least in the next section) both rely on a self-clock that is at least
partially derived from the properties of the subnet under test. This partially derived from the properties of the subnet under test. This
introduces the possibility that the subnet under test exhibits introduces the possibility that the subnet under test exhibits
behaviors such as extreme RTT fluctuations that prevent these behaviors such as extreme RTT fluctuations that prevent these
algorithms from accurately controlling data rates. algorithms from accurately controlling data rates.
A FS-TIDS specifying a constant window CBR test must explicitly An FSTIDS specifying a Constant Window Pseudo CBR test must
indicate under what conditions errors in the data rate cause tests to explicitly indicate under what conditions errors in the data rate
be inconclusive. Conventional paced measurement traffic may be more cause tests to be inconclusive. Conventional paced measurement
appropriate for these environments. traffic may be more appropriate for these environments.
6.3. Scanned window pseudo CBR 6.3. Scanned Window Pseudo CBR
Scanned window pseudo CBR is similar to the constant window CBR Scanned Window Pseudo CBR is similar to the Constant Window Pseudo
described above, except the window is scanned across a range of sizes CBR described above, except the window is scanned across a range of
designed to include two key events, the onset of queuing and the sizes designed to include two key events: the onset of queuing and
onset of packet loss or ECN CE marks. The window is scanned by the onset of packet loss or ECN CE marks. The window is scanned by
incrementing it by one packet every 2*target_window_size delivered incrementing it by one packet every 2*target_window_size delivered
packets. This mimics the additive increase phase of standard Reno packets. This mimics the additive increase phase of standard Reno
TCP congestion avoidance when delayed ACKs are in effect. Normally TCP congestion avoidance when delayed ACKs are in effect. Normally,
the window increases separated by intervals slightly longer than the window increases are separated by intervals slightly longer than
twice the target_RTT. twice the target_RTT.
There are two ways to implement this test: one built by applying a There are two ways to implement this test: 1) applying a window clamp
window clamp to standard congestion control in a standard protocol to standard congestion control in a standard protocol such as TCP and
such as TCP and the other built by stiffening a non-standard 2) stiffening a non-standard transport protocol. When standard
transport protocol. When standard congestion control is in effect, congestion control is in effect, any losses or ECN CE marks cause the
any losses or ECN CE marks cause the transport to revert to a window transport to revert to a window smaller than the clamp, such that the
smaller than the clamp such that the scanning clamp loses control the scanning clamp loses control of the window size. The NPAD (Network
window size. The NPAD pathdiag tool is an example of this class of Path and Application Diagnostics) pathdiag tool is an example of this
algorithms [Pathdiag]. class of algorithms [Pathdiag].
Alternatively a non-standard congestion control algorithm can respond Alternatively, a non-standard congestion control algorithm can
to losses by transmitting extra data, such that it maintains the respond to losses by transmitting extra data, such that it maintains
specified window size independent of losses or ECN CE marks. Such a the specified window size independent of losses or ECN CE marks.
stiffened transport explicitly violates mandatory Internet congestion Such a stiffened transport explicitly violates mandatory Internet
control [RFC5681] and is not suitable for in situ testing. It is congestion control [RFC5681] and is not suitable for in situ testing.
only appropriate for engineering testing under laboratory conditions. It is only appropriate for engineering testing under laboratory
The Windowed Ping tool implements such a test [WPING]. The tool conditions. The Windowed Ping tool implements such a test [WPING].
described in the paper has been updated.[mpingSource] This tool has been updated (see [mpingSource]).
The test procedures in Section 8.2 describe how to the partition the The test procedures in Section 8.2 describe how to the partition the
scans into regions and how to interpret the results. scans into regions and how to interpret the results.
6.4. Concurrent or channelized testing 6.4. Concurrent or Channelized Testing
The procedures described in this document are only directly The procedures described in this document are only directly
applicable to single stream measurement, e.g. one TCP connection or applicable to single-stream measurement, e.g., one TCP connection or
measurement stream. In an ideal world, we would disallow all measurement stream. In an ideal world, we would disallow all
performance claims based multiple concurrent streams, but this is not performance claims based on multiple concurrent streams, but this is
practical due to at least two issues. First, many very high rate not practical due to at least two issues. First, many very high-rate
link technologies are channelized and at last partially pin the flow link technologies are channelized and at last partially pin the flow-
to channel mapping to minimize packet reordering within flows. to-channel mapping to minimize packet reordering within flows.
Second, TCP itself has scaling limits. Although the former problem Second, TCP itself has scaling limits. Although the former problem
might be overcome through different design decisions, the later might be overcome through different design decisions, the latter
problem is more deeply rooted. problem is more deeply rooted.
All congestion control algorithms that are philosophically aligned All congestion control algorithms that are philosophically aligned
with the standard [RFC5681] (e.g. claim some level of TCP with [RFC5681] (e.g., claim some level of TCP compatibility,
compatibility, friendliness or fairness) have scaling limits, in the friendliness, or fairness) have scaling limits; that is, as a long
sense that as a long fast network (LFN) with a fixed RTT and MTU gets fat network (LFN) with a fixed RTT and MTU gets faster, these
faster, these congestion control algorithms get less accurate and as congestion control algorithms get less accurate and, as a
a consequence have difficulty filling the network [CCscaling]. These consequence, have difficulty filling the network [CCscaling]. These
properties are a consequence of the original Reno AIMD congestion properties are a consequence of the original Reno AIMD congestion
control design and the requirement in [RFC5681] that all transport control design and the requirement in [RFC5681] that all transport
protocols have similar responses to congestion. protocols have similar responses to congestion.
There are a number of reasons to want to specify performance in terms There are a number of reasons to want to specify performance in terms
of multiple concurrent flows, however this approach is not of multiple concurrent flows; however, this approach is not
recommended for data rates below several megabits per second, which recommended for data rates below several megabits per second, which
can be attained with run lengths under 10000 packets on many paths. can be attained with run lengths under 10000 packets on many paths.
Since the required run length goes as the square of the data rate, at Since the required run length is proportional to the square of the
higher rates the run lengths can be unreasonably large, and multiple data rate, at higher rates, the run lengths can be unreasonably
flows might be the only feasible approach. large, and multiple flows might be the only feasible approach.
If multiple flows are deemed necessary to meet aggregate performance If multiple flows are deemed necessary to meet aggregate performance
targets then this must be stated in both the design of the TIDS and targets, then this must be stated both in the design of the TIDS and
in any claims about network performance. The IP diagnostic tests in any claims about network performance. The IP diagnostic tests
must be performed concurrently with the specified number of must be performed concurrently with the specified number of
connections. For the tests that use bursty test streams, the bursts connections. For the tests that use bursty test streams, the bursts
should be synchronized across streams unless there is a priori should be synchronized across streams unless there is a priori
knowledge that the applications have some explicit mechanism to knowledge that the applications have some explicit mechanism to
stagger their own bursts. In the absences of an explicit mechanism stagger their own bursts. In the absence of an explicit mechanism to
to stagger bursts many network and application artifacts will stagger bursts, many network and application artifacts will sometimes
sometimes implicitly synchronize bursts. A test that does not implicitly synchronize bursts. A test that does not control burst
control burst synchronization may be prone to false pass results for synchronization may be prone to false pass results for some
some applications. applications.
7. Interpreting the Results 7. Interpreting the Results
7.1. Test outcomes 7.1. Test Outcomes
To perform an exhaustive test of a complete network path, each test To perform an exhaustive test of a complete network path, each test
of the TIDS is applied to each subpath of the complete path. If any of the TIDS is applied to each subpath of the complete path. If any
subpath fails any test then a standard transport protocol running subpath fails any test, then a standard transport protocol running
over the complete path can also be expected to fail to attain the over the complete path can also be expected to fail to attain the
Target Transport Performance under some conditions. Target Transport Performance under some conditions.
In addition to passing or failing, a test can be deemed to be In addition to passing or failing, a test can be deemed to be
inconclusive for a number of reasons. Proper instrumentation and inconclusive for a number of reasons. Proper instrumentation and
treatment of inconclusive outcomes is critical to the accuracy and treatment of inconclusive outcomes is critical to the accuracy and
robustness of Model Based Metrics. Tests can be inconclusive if the robustness of Model-Based Metrics. Tests can be inconclusive if the
precomputed traffic pattern or data rates were not accurately precomputed traffic pattern or data rates were not accurately
generated; the measurement results were not statistically generated; the measurement results were not statistically
significant; and others causes such as failing to meet some required significant; the required preconditions for the test were not met; or
preconditions for the test. See Section 5.4 other causes. See Section 5.4.
For example consider a test that implements Constant Window Pseudo For example, consider a test that implements Constant Window Pseudo
CBR (Section 6.2) by adding rate controls and detailed IP packet CBR (Section 6.2) by adding rate controls and detailed IP packet
transfer instrumentation to TCP (e.g. [RFC4898]). TCP includes transfer instrumentation to TCP (e.g., using the extended performance
built in control systems which might interfere with the sending data statistics for TCP as described in [RFC4898]). TCP includes built-in
rate. If such a test meets the required packet transfer statistics control systems that might interfere with the sending data rate. If
(e.g. run length) while failing to attain the specified data rate it such a test meets the required packet transfer statistics (e.g., run
must be treated as an inconclusive result, because we can not a length) while failing to attain the specified data rate, it must be
priori determine if the reduced data rate was caused by a TCP problem treated as an inconclusive result, because we cannot a priori
or a network problem, or if the reduced data rate had a material determine if the reduced data rate was caused by a TCP problem or a
effect on the observed packet transfer statistics. network problem or if the reduced data rate had a material effect on
the observed packet transfer statistics.
Note that for capacity tests, if the observed packet transfer Note that for capacity tests, if the observed packet transfer
statistics meet the statistical criteria for failing (accepting statistics meet the statistical criteria for failing (based on
hypnosis H1 in Section 7.2), the test can can be considered to have acceptance of hypothesis H1 in Section 7.2), the test can be
failed because it doesn't really matter that the test didn't attain considered to have failed because it doesn't really matter that the
the required data rate. test didn't attain the required data rate.
The really important new properties of MBM, such as vantage The important new properties of MBM, such as vantage independence,
independence, are a direct consequence of opening the control loops are a direct consequence of opening the control loops in the
in the protocols, such that the test stream does not depend on protocols, such that the test stream does not depend on network
network conditions or IP packets received. Any mechanism that conditions or IP packets received. Any mechanism that introduces
introduces feedback between the path's measurements and the test feedback between the path's measurements and the test stream
stream generation is at risk of introducing nonlinearities that spoil generation is at risk of introducing nonlinearities that spoil these
these properties. Any exceptional event that indicates that such properties. Any exceptional event that indicates that such feedback
feedback has happened should cause the test to be considered has happened should cause the test to be considered inconclusive.
inconclusive.
One way to view inconclusive tests is that they reflect situations Inconclusive tests may be caused by situations in which a test
where a test outcome is ambiguous between limitations of the network outcome is ambiguous because of network limitations or an unknown
and some unknown limitation of the IP diagnostic test itself, which limitation on the IP diagnostic test itself, which may have been
may have been caused by some uncontrolled feedback from the network. caused by some uncontrolled feedback from the network.
Note that procedures that attempt to search the target parameter Note that procedures that attempt to search the target parameter
space to find the limits on some parameter such as target_data_rate space to find the limits on a parameter such as target_data_rate are
are at risk of breaking the location independent properties of Model at risk of breaking the location-independent properties of Model-
Based Metrics, if any part of the boundary between passing and Based Metrics if any part of the boundary between passing,
inconclusive or failing results is sensitive to RTT (which is inconclusive, or failing results is sensitive to RTT (which is
normally the case). For example the maximum data rate for a marginal normally the case). For example, the maximum data rate for a
link (e.g. exhibiting excess errors) is likely to be sensitive to marginal link (e.g., exhibiting excess errors) is likely to be
the test_path_RTT. The maximum observed data rate over the test path sensitive to the test_path_RTT. The maximum observed data rate over
has very little value for predicting the maximum rate over a the test path has very little value for predicting the maximum rate
different path. over a different path.
One of the goals for evolving TIDS designs will be to keep sharpening One of the goals for evolving TIDS designs will be to keep sharpening
distinction between inconclusive, passing and failing tests. The the distinctions between inconclusive, passing, and failing tests.
criteria for for passing, failing and inconclusive tests must be The criteria for inconclusive, passing, and failing tests must be
explicitly stated for every test in the TIDS or FS-TIDS. explicitly stated for every test in the TIDS or FSTIDS.
One of the goals of evolving the testing process, procedures, tools One of the goals for evolving the testing process, procedures, tools,
and measurement point selection should be to minimize the number of and measurement point selection should be to minimize the number of
inconclusive tests. inconclusive tests.
It may be useful to keep raw packet transfer statistics and ancillary It may be useful to keep raw packet transfer statistics and ancillary
metrics [RFC3148] for deeper study of the behavior of the network metrics [RFC3148] for deeper study of the behavior of the network
path and to measure the tools themselves. Raw packet transfer path and to measure the tools themselves. Raw packet transfer
statistics can help to drive tool evolution. Under some conditions statistics can help to drive tool evolution. Under some conditions,
it might be possible to re-evaluate the raw data for satisfying it might be possible to re-evaluate the raw data for satisfying
alternate Target Transport Performance. However it is important to alternate Target Transport Performance. However, it is important to
guard against sampling bias and other implicit feedback which can guard against sampling bias and other implicit feedback that can
cause false results and exhibit measurement point vantage cause false results and exhibit measurement point vantage
sensitivity. Simply applying different delivery criteria based on a sensitivity. Simply applying different delivery criteria based on a
different Target Transport Performance is insufficient if the test different Target Transport Performance is insufficient if the test
traffic patterns (bursts, etc.) does not match the alternate Target traffic patterns (bursts, etc.) do not match the alternate Target
Transport Performance. Transport Performance.
7.2. Statistical criteria for estimating run_length 7.2. Statistical Criteria for Estimating run_length
When evaluating the observed run_length, we need to determine When evaluating the observed run_length, we need to determine
appropriate packet stream sizes and acceptable error levels for appropriate packet stream sizes and acceptable error levels for
efficient measurement. In practice, can we compare the empirically efficient measurement. In practice, can we compare the empirically
estimated packet loss and ECN Congestion Experienced (CE) marking estimated packet loss and ECN CE marking ratios with the targets as
ratios with the targets as the sample size grows? How large a sample the sample size grows? How large a sample is needed to say that the
is needed to say that the measurements of packet transfer indicate a measurements of packet transfer indicate a particular run length is
particular run length is present? present?
The generalized measurement can be described as recursive testing: The generalized measurement can be described as recursive testing:
send packets (individually or in patterns) and observe the packet send packets (individually or in patterns) and observe the packet
transfer performance (packet loss ratio or other metric, any marking transfer performance (packet loss ratio, other metric, or any marking
we define). we define).
As each packet is sent and measured, we have an ongoing estimate of As each packet is sent and measured, we have an ongoing estimate of
the performance in terms of the ratio of packet loss or ECN CE mark the performance in terms of the ratio of packet loss or ECN CE marks
to total packets (i.e. an empirical probability). We continue to to total packets (i.e., an empirical probability). We continue to
send until conditions support a conclusion or a maximum sending limit send until conditions support a conclusion or a maximum sending limit
has been reached. has been reached.
We have a target_mark_probability, 1 mark per target_run_length, We have a target_mark_probability, one mark per target_run_length,
where a "mark" is defined as a lost packet, a packet with ECN CE where a "mark" is defined as a lost packet, a packet with ECN CE
mark, or other signal. This constitutes the null Hypothesis: mark, or other signal. This constitutes the null hypothesis:
H0: no more than one mark in target_run_length = H0: no more than one mark in target_run_length =
3*(target_window_size)^2 packets 3*(target_window_size)^2 packets
and we can stop sending packets if on-going measurements support We can stop sending packets if ongoing measurements support accepting
accepting H0 with the specified Type I error = alpha (= 0.05 for H0 with the specified Type I error = alpha (= 0.05, for example).
example).
We also have an alternative Hypothesis to evaluate: if performance is We also have an alternative hypothesis to evaluate: is performance
significantly lower than the target_mark_probability. Based on significantly lower than the target_mark_probability? Based on
analysis of typical values and practical limits on measurement analysis of typical values and practical limits on measurement
duration, we choose four times the H0 probability: duration, we choose four times the H0 probability:
H1: one or more marks in (target_run_length/4) packets H1: one or more marks in (target_run_length/4) packets
and we can stop sending packets if measurements support rejecting H0 and we can stop sending packets if measurements support rejecting H0
with the specified Type II error = beta (= 0.05 for example), thus with the specified Type II error = beta (= 0.05, for example), thus
preferring the alternate hypothesis H1. preferring the alternate hypothesis H1.
H0 and H1 constitute the Success and Failure outcomes described H0 and H1 constitute the success and failure outcomes described
elsewhere in the memo, and while the ongoing measurements do not elsewhere in this document; while the ongoing measurements do not
support either hypothesis the current status of measurements is support either hypothesis, the current status of measurements is
inconclusive. inconclusive.
The problem above is formulated to match the Sequential Probability The problem above is formulated to match the Sequential Probability
Ratio Test (SPRT) [Wald45] and [Montgomery90]. Note that as Ratio Test (SPRT) [Wald45] [Montgomery90]. Note that as originally
originally framed the events under consideration were all framed, the events under consideration were all manufacturing
manufacturing defects. In networking, ECN CE marks and lost packets defects. In networking, ECN CE marks and lost packets are not
are not defects but signals, indicating that the transport protocol defects but signals, indicating that the transport protocol should
should slow down. slow down.
The Sequential Probability Ratio Test also starts with a pair of The Sequential Probability Ratio Test also starts with a pair of
hypothesis specified as above: hypotheses specified as above:
H0: p0 = one defect in target_run_length H0: p0 = one defect in target_run_length
H1: p1 = one defect in target_run_length/4 H1: p1 = one defect in target_run_length/4
As packets are sent and measurements collected, the tester evaluates As packets are sent and measurements collected, the tester evaluates
the cumulative defect count against two boundaries representing H0 the cumulative defect count against two boundaries representing H0
Acceptance or Rejection (and acceptance of H1): Acceptance or Rejection (and acceptance of H1):
Acceptance line: Xa = -h1 + s*n Acceptance line: Xa = -h1 + s*n
Rejection line: Xr = h2 + s*n Rejection line: Xr = h2 + s*n
where n increases linearly for each packet sent and where n increases linearly for each packet sent and
h1 = { log((1-alpha)/beta) }/k h1 = { log((1-alpha)/beta) }/k
h2 = { log((1-beta)/alpha) }/k h2 = { log((1-beta)/alpha) }/k
k = log{ (p1(1-p0)) / (p0(1-p1)) } k = log{ (p1(1-p0)) / (p0(1-p1)) }
s = [ log{ (1-p0)/(1-p1) } ]/k s = [ log{ (1-p0)/(1-p1) } ]/k
for p0 and p1 as defined in the null and alternative Hypotheses for p0 and p1 as defined in the null and alternative hypotheses
statements above, and alpha and beta as the Type I and Type II statements above, and alpha and beta as the Type I and Type II
errors. errors.
The SPRT specifies simple stopping rules: The SPRT specifies simple stopping rules:
o Xa < defect_count(n) < Xr: continue testing o Xa < defect_count(n) < Xr: continue testing
o defect_count(n) <= Xa: Accept H0 o defect_count(n) <= Xa: Accept H0
o defect_count(n) >= Xr: Accept H1 o defect_count(n) >= Xr: Accept H1
The calculations above are implemented in the R-tool for Statistical The calculations above are implemented in the R-tool for Statistical
Analysis [Rtool] , in the add-on package for Cross-Validation via Analysis [Rtool], in the add-on package for Cross-Validation via
Sequential Testing (CVST) [CVST]. Sequential Testing (CVST) [CVST].
Using the equations above, we can calculate the minimum number of Using the equations above, we can calculate the minimum number of
packets (n) needed to accept H0 when x defects are observed. For packets (n) needed to accept H0 when x defects are observed. For
example, when x = 0: example, when x = 0:
Xa = 0 = -h1 + s*n Xa = 0 = -h1 + s*n
and n = h1 / s and n = h1 / s
Note that the derivations in [Wald45] and [Montgomery90] differ. Note that the derivations in [Wald45] and [Montgomery90] differ.
Montgomery's simplified derivation of SPRT may assume a Bernoulli Montgomery's simplified derivation of SPRT may assume a Bernoulli
processes, where the packet loss probabilities are independent and processes, where the packet loss probabilities are independent and
identically distributed, making the SPRT more accessible. Wald's identically distributed, making the SPRT more accessible. Wald's
seminal paper showed that this assumption is not necessary. It helps seminal paper showed that this assumption is not necessary. It helps
to remember that the goal of SPRT is not to estimate the value of the to remember that the goal of SPRT is not to estimate the value of the
packet loss rate, but only whether or not the packet loss ratio is packet loss rate but only whether or not the packet loss ratio is
likely low enough (when we accept the H0 null hypothesis) yielding likely (1) low enough (when we accept the H0 null hypothesis),
success; or too high (when we accept the H1 alternate hypothesis) yielding success or (2) too high (when we accept the H1 alternate
yielding failure. hypothesis), yielding failure.
7.3. Reordering Tolerance 7.3. Reordering Tolerance
All tests must be instrumented for packet level reordering [RFC4737]. All tests must be instrumented for packet-level reordering [RFC4737].
However, there is no consensus for how much reordering should be However, there is no consensus for how much reordering should be
acceptable. Over the last two decades the general trend has been to acceptable. Over the last two decades, the general trend has been to
make protocols and applications more tolerant to reordering (see for make protocols and applications more tolerant to reordering (for
example [RFC4015]), in response to the gradual increase in reordering example, see [RFC5827]), in response to the gradual increase in
in the network. This increase has been due to the deployment of reordering in the network. This increase has been due to the
technologies such as multithreaded routing lookups and Equal Cost deployment of technologies such as multithreaded routing lookups and
MultiPath (ECMP) routing. These techniques increase parallelism in Equal-Cost Multipath (ECMP) routing. These techniques increase
network and are critical to enabling overall Internet growth to parallelism in the network and are critical to enabling overall
exceed Moore's Law. Internet growth to exceed Moore's Law.
Note that transport retransmission strategies can trade off With transport retransmission strategies, there are fundamental
reordering tolerance vs how quickly they can repair losses vs trade-offs among reordering tolerance, how quickly losses can be
overhead from spurious retransmissions. In advance of new repaired, and overhead from spurious retransmissions. In advance of
retransmission strategies we propose the following strawman: new retransmission strategies, we propose the following strawman:
Transport protocols should be able to adapt to reordering as long as transport protocols should be able to adapt to reordering as long as
the reordering extent is not more than the maximum of one quarter the reordering extent is not more than the maximum of one quarter
window or 1 mS, whichever is larger. (These values come from window or 1 ms, whichever is larger. (These values come from
experience prototyping Early Retransmit [RFC5827] and related experience prototyping Early Retransmit [RFC5827] and related
algorithms. They agree with the values being proposed for "RACK: a algorithms. They agree with the values being proposed for "RACK: a
time-based fast loss detection algorithm" [I-D.ietf-tcpm-rack].) time-based fast loss detection algorithm" [RACK].) Within this limit
Within this limit on reorder extent, there should be no bound on on reorder extent, there should be no bound on reordering density.
reordering density.
By implication, recording which is less than these bounds should not By implication, recording that is less than these bounds should not
be treated as a network impairment. However [RFC4737] still applies: be treated as a network impairment. However, [RFC4737] still
reordering should be instrumented and the maximum reordering that can applies: reordering should be instrumented, and the maximum
be properly characterized by the test (because of the bound on reordering that can be properly characterized by the test (because of
history buffers) should be recorded with the measurement results. the bound on history buffers) should be recorded with the measurement
results.
Reordering tolerance and diagnostic limitations, such as the size of Reordering tolerance and diagnostic limitations, such as the size of
the history buffer used to diagnose packets that are way out-of- the history buffer used to diagnose packets that are way out of
order, must be specified in a FSTIDS. order, must be specified in an FSTIDS.
8. IP Diagnostic Tests 8. IP Diagnostic Tests
The IP diagnostic tests below are organized according to the The IP diagnostic tests below are organized according to the
technique used to generate the test stream as described in Section 6. technique used to generate the test stream as described in Section 6.
All of the results are evaluated in accordance with Section 7, All of the results are evaluated in accordance with Section 7,
possibly with additional test specific critera. possibly with additional test-specific criteria.
We also introduce some combined tests which are more efficient when We also introduce some combined tests that are more efficient when
networks are expected to pass, but conflate diagnostic signatures networks are expected to pass but conflate diagnostic signatures when
when they fail. they fail.
8.1. Basic Data Rate and Packet Transfer Tests 8.1. Basic Data Rate and Packet Transfer Tests
We propose several versions of the basic data rate and packet We propose several versions of the basic data rate and packet
transfer statistics test that differ in how the data rate is transfer statistics test that differ in how the data rate is
controlled. The data can be paced on a timer, or window controlled controlled. The data can be paced on a timer or window controlled
(and self clocked). The first two tests implicitly confirm that (and self-clocked). The first two tests implicitly confirm that
sub_path has sufficient raw capacity to carry the target_data_rate. sub_path has sufficient raw capacity to carry the target_data_rate.
They are recommended for relatively infrequent testing, such as an They are recommended for relatively infrequent testing, such as an
installation or periodic auditing process. The third, background installation or periodic auditing process. The third test,
packet transfer statistics, is a low rate test designed for ongoing Background Packet Transfer Statistics, is a low-rate test designed
monitoring for changes in subpath quality. for ongoing monitoring for changes in subpath quality.
8.1.1. Delivery Statistics at Paced Full Data Rate 8.1.1. Delivery Statistics at Paced Full Data Rate
Confirm that the observed run length is at least the This test confirms that the observed run length is at least the
target_run_length while relying on timer to send data at the target_run_length while relying on timer to send data at the
target_rate using the procedure described in in Section 6.1 with a target_rate using the procedure described in Section 6.1 with a burst
burst size of 1 (single packets) or 2 (packet pairs). size of 1 (single packets) or 2 (packet pairs).
The test is considered to be inconclusive if the packet transmission The test is considered to be inconclusive if the packet transmission
can not be accurately controlled for any reason. cannot be accurately controlled for any reason.
RFC 6673 [RFC6673] is appropriate for measuring packet transfer RFC 6673 [RFC6673] is appropriate for measuring packet transfer
statistics at full data rate. statistics at full data rate.
8.1.2. Delivery Statistics at Full Data Windowed Rate 8.1.2. Delivery Statistics at Full Data Windowed Rate
Confirm that the observed run length is at least the This test confirms that the observed run length is at least the
target_run_length while sending at an average rate approximately target_run_length while sending at an average rate approximately
equal to the target_data_rate, by controlling (or clamping) the equal to the target_data_rate, by controlling (or clamping) the
window size of a conventional transport protocol to test_window. window size of a conventional transport protocol to test_window.
Since losses and ECN CE marks cause transport protocols to reduce Since losses and ECN CE marks cause transport protocols to reduce
their data rates, this test is expected to be less precise about their data rates, this test is expected to be less precise about
controlling its data rate. It should not be considered inconclusive controlling its data rate. It should not be considered inconclusive
as long as at least some of the round trips reached the full as long as at least some of the round trips reached the full
target_data_rate without incurring losses or ECN CE marks. To pass target_data_rate without incurring losses or ECN CE marks. To pass
this test the network must deliver target_window_size packets in this test, the network must deliver target_window_size packets in
target_RTT time without any losses or ECN CE marks at least once per target_RTT time without any losses or ECN CE marks at least once per
two target_window_size round trips, in addition to meeting the run two target_window_size round trips, in addition to meeting the run
length statistical test. length statistical test.
8.1.3. Background Packet Transfer Statistics Tests 8.1.3. Background Packet Transfer Statistics Tests
The background run length is a low rate version of the target target The Background Packet Transfer Statistics Test is a low-rate version
rate test above, designed for ongoing lightweight monitoring for of the target rate test above, designed for ongoing lightweight
changes in the observed subpath run length without disrupting users. monitoring for changes in the observed subpath run length without
It should be used in conjunction with one of the above full rate disrupting users. It should be used in conjunction with one of the
tests because it does not confirm that the subpath can support raw above full-rate tests because it does not confirm that the subpath
data rate. can support raw data rate.
RFC 6673 [RFC6673] is appropriate for measuring background packet RFC 6673 [RFC6673] is appropriate for measuring background packet
transfer statistics. transfer statistics.
8.2. Standing Queue Tests 8.2. Standing Queue Tests
These engineering tests confirm that the bottleneck is well behaved These engineering tests confirm that the bottleneck is well behaved
across the onset of packet loss, which typically follows after the across the onset of packet loss, which typically follows after the
onset of queuing. Well behaved generally means lossless for onset of queuing. Well behaved generally means lossless for
transient queues, but once the queue has been sustained for a transient queues, but once the queue has been sustained for a
sufficient period of time (or reaches a sufficient queue depth) there sufficient period of time (or reaches a sufficient queue depth),
should be a small number of losses or ECN CE marks to signal to the there should be a small number of losses or ECN CE marks to signal to
transport protocol that it should reduce its window or data rate. the transport protocol that it should reduce its window or data rate.
Losses that are too early can prevent the transport from averaging at Losses that are too early can prevent the transport from averaging at
the target_data_rate. Losses that are too late indicate that the the target_data_rate. Losses that are too late indicate that the
queue might not have an appropriate AQM [RFC7567] and as a queue might not have an appropriate AQM [RFC7567] and, as a
consequence subject to bufferbloat [wikiBloat]. Queues without AQM consequence, be subject to bufferbloat [wikiBloat]. Queues without
have the potential to inflict excess delays on all flows sharing the AQM have the potential to inflict excess delays on all flows sharing
bottleneck. Excess losses (more than half of the window) at the the bottleneck. Excess losses (more than half of the window) at the
onset of loss make loss recovery problematic for the transport onset of loss make loss recovery problematic for the transport
protocol. Non-linear, erratic or excessive RTT increases suggest protocol. Non-linear, erratic, or excessive RTT increases suggest
poor interactions between the channel acquisition algorithms and the poor interactions between the channel acquisition algorithms and the
transport self clock. All of the tests in this section use the same transport self-clock. All of the tests in this section use the same
basic scanning algorithm, described here, but score the link or basic scanning algorithm, described here, but score the link or
subpath on the basis of how well it avoids each of these problems. subpath on the basis of how well it avoids each of these problems.
Some network technologies rely on virtual queues or other techniques Some network technologies rely on virtual queues or other techniques
to meter traffic without adding any queuing delay, in which case the to meter traffic without adding any queuing delay, in which case the
data rate will vary with the window size all the way up to the onset data rate will vary with the window size all the way up to the onset
of load induced packet loss or ECN CE marks. For these technologies, of load-induced packet loss or ECN CE marks. For these technologies,
the discussion of queuing in Section 6.3 does not apply, but it is the discussion of queuing in Section 6.3 does not apply, but it is
still necessary to confirm that the onset of losses or ECN CE marks still necessary to confirm that the onset of losses or ECN CE marks
be at an appropriate point and progressive. If the network be at an appropriate point and progressive. If the network
bottleneck does not introduce significant queuing delay, modify the bottleneck does not introduce significant queuing delay, modify the
procedure described in Section 6.3 to start the scan at a window procedure described in Section 6.3 to start the scan at a window
equal to or slightly smaller than the test_window. equal to or slightly smaller than the test_window.
Use the procedure in Section 6.3 to sweep the window across the onset Use the procedure in Section 6.3 to sweep the window across the onset
of queuing and the onset of loss. The tests below all assume that of queuing and the onset of loss. The tests below all assume that
the scan emulates standard additive increase and delayed ACK by the scan emulates standard additive increase and delayed ACK by
incrementing the window by one packet for every 2*target_window_size incrementing the window by one packet for every 2*target_window_size
packets delivered. A scan can typically be divided into three packets delivered. A scan can typically be divided into three
regions: below the onset of queuing, a standing queue, and at or regions: below the onset of queuing, a standing queue, and at or
beyond the onset of loss. beyond the onset of loss.
Below the onset of queuing the RTT is typically fairly constant, and Below the onset of queuing, the RTT is typically fairly constant, and
the data rate varies in proportion to the window size. Once the data the data rate varies in proportion to the window size. Once the data
rate reaches the subpath IP rate, the data rate becomes fairly rate reaches the subpath IP rate, the data rate becomes fairly
constant, and the RTT increases in proportion to the increase in constant, and the RTT increases in proportion to the increase in
window size. The precise transition across the start of queuing can window size. The precise transition across the start of queuing can
be identified by the maximum network power, defined to be the ratio be identified by the maximum network power, defined to be the ratio
data rate over the RTT. The network power can be computed at each data rate over the RTT. The network power can be computed at each
window size, and the window with the maximum is taken as the start of window size, and the window with the maximum is taken as the start of
the queuing region. the queuing region.
If there is random background loss (e.g. bit errors, etc), precise If there is random background loss (e.g., bit errors), precise
determination of the onset of queue induced packet loss may require determination of the onset of queue-induced packet loss may require
multiple scans. Above the onset of queuing loss, all transport multiple scans. At window sizes large enough to cause loss in
protocols are expected to experience periodic losses determined by queues, all transport protocols are expected to experience periodic
the interaction between the congestion control and AQM algorithms. losses determined by the interaction between the congestion control
For standard congestion control algorithms the periodic losses are and AQM algorithms. For standard congestion control algorithms, the
likely to be relatively widely spaced and the details are typically periodic losses are likely to be relatively widely spaced, and the
dominated by the behavior of the transport protocol itself. For the details are typically dominated by the behavior of the transport
stiffened transport protocols case (with non-standard, aggressive protocol itself. For the case of stiffened transport protocols (with
congestion control algorithms) the details of periodic losses will be non-standard, aggressive congestion control algorithms), the details
dominated by how the window increase function responds to loss. of periodic losses will be dominated by how the window increase
function responds to loss.
8.2.1. Congestion Avoidance 8.2.1. Congestion Avoidance
A subpath passes the congestion avoidance standing queue test if more A subpath passes the congestion avoidance standing queue test if more
than target_run_length packets are delivered between the onset of than target_run_length packets are delivered between the onset of
queuing (as determined by the window with the maximum network power queuing (as determined by the window with the maximum network power
as described above) and the first loss or ECN CE mark. If this test as described above) and the first loss or ECN CE mark. If this test
is implemented using a standard congestion control algorithm with a is implemented using a standard congestion control algorithm with a
clamp, it can be performed in situ in the production internet as a clamp, it can be performed in situ in the production internet as a
capacity test. For an example of such a test see [Pathdiag]. capacity test. For an example of such a test, see [Pathdiag].
For technologies that do not have conventional queues, use the For technologies that do not have conventional queues, use the
test_window in place of the onset of queuing. i.e. A subpath passes test_window in place of the onset of queuing. That is, a subpath
the congestion avoidance standing queue test if more than passes the congestion avoidance standing queue test if more than
target_run_length packets are delivered between start of the scan at target_run_length packets are delivered between the start of the scan
test_window and the first loss or ECN CE mark. at test_window and the first loss or ECN CE mark.
8.2.2. Bufferbloat 8.2.2. Bufferbloat
This test confirms that there is some mechanism to limit buffer This test confirms that there is some mechanism to limit buffer
occupancy (e.g. that prevents bufferbloat). Note that this is not occupancy (e.g., that prevents bufferbloat). Note that this is not
strictly a requirement for single stream bulk transport capacity, strictly a requirement for single-stream bulk transport capacity;
however if there is no mechanism to limit buffer queue occupancy then however, if there is no mechanism to limit buffer queue occupancy,
a single stream with sufficient data to deliver is likely to cause then a single stream with sufficient data to deliver is likely to
the problems described in [RFC7567], and [wikiBloat]. This may cause cause the problems described in [RFC7567] and [wikiBloat]. This may
only minor symptoms for the dominant flow, but has the potential to cause only minor symptoms for the dominant flow but has the potential
make the subpath unusable for other flows and applications. to make the subpath unusable for other flows and applications.
Pass if the onset of loss occurs before a standing queue has The test will pass if the onset of loss occurs before a standing
introduced more delay than than twice target_RTT, or other well queue has introduced delay greater than twice the target_RTT or
defined and specified limit. Note that there is not yet a model for another well-defined and specified limit. Note that there is not yet
how much standing queue is acceptable. The factor of two chosen here a model for how much standing queue is acceptable. The factor of two
reflects a rule of thumb. In conjunction with the previous test, chosen here reflects a rule of thumb. In conjunction with the
this test implies that the first loss should occur at a queuing delay previous test, this test implies that the first loss should occur at
which is between one and two times the target_RTT. a queuing delay that is between one and two times the target_RTT.
Specified RTT limits that are larger than twice the target_RTT must Specified RTT limits that are larger than twice the target_RTT must
be fully justified in the FS-TIDS. be fully justified in the FSTIDS.
8.2.3. Non excessive loss 8.2.3. Non-excessive Loss
This test confirms that the onset of loss is not excessive. Pass if This test confirms that the onset of loss is not excessive. The test
losses are equal or less than the increase in the cross traffic plus will pass if losses are equal to or less than the increase in the
the test stream window increase since the previous RTT. This could cross traffic plus the test stream window increase since the previous
be restated as non-decreasing total throughput of the subpath at the RTT. This could be restated as non-decreasing total throughput of
onset of loss. (Note that when there is a transient drop in subpath the subpath at the onset of loss. (Note that when there is a
throughput and there is not already a standing queue, a subpath that transient drop in subpath throughput and there is not already a
passes other queue tests in this document will have sufficient queue standing queue, a subpath that passes other queue tests in this
space to hold one full RTT worth of data). document will have sufficient queue space to hold one full RTT worth
of data).
Note that token bucket policers will not pass this test, which is as Note that token bucket policers will not pass this test, which is as
intended. TCP often stumbles badly if more than a small fraction of intended. TCP often stumbles badly if more than a small fraction of
the packets are dropped in one RTT. Many TCP implementations will the packets are dropped in one RTT. Many TCP implementations will
require a timeout and slowstart to recover their self clock. Even if require a timeout and slowstart to recover their self-clock. Even if
they can recover from the massive losses the sudden change in they can recover from the massive losses, the sudden change in
available capacity at the bottleneck wastes serving and front path available capacity at the bottleneck wastes serving and front-path
capacity until TCP can adapt to the new rate [Policing]. capacity until TCP can adapt to the new rate [Policing].
8.2.4. Duplex Self Interference 8.2.4. Duplex Self-Interference
This engineering test confirms a bound on the interactions between This engineering test confirms a bound on the interactions between
the forward data path and the ACK return path when they share a half the forward data path and the ACK return path when they share a half-
duplex link. duplex link.
Some historical half duplex technologies had the property that each Some historical half-duplex technologies had the property that each
direction held the channel until it completely drained its queue. direction held the channel until it completely drained its queue.
When a self clocked transport protocol, such as TCP, has data and When a self-clocked transport protocol, such as TCP, has data and
ACKs passing in opposite directions through such a link, the behavior ACKs passing in opposite directions through such a link, the behavior
often reverts to stop-and-wait. Each additional packet added to the often reverts to stop-and-wait. Each additional packet added to the
window raises the observed RTT by two packet times, once as the window raises the observed RTT by two packet times, once as the
additional packet passes through the data path, and once for the additional packet passes through the data path and once for the
additional delay incurred by the ACK waiting on the return path. additional delay incurred by the ACK waiting on the return path.
The duplex self interference test fails if the RTT rises by more than The Duplex Self-Interference Test fails if the RTT rises by more than
a fixed bound above the expected queuing time computed from the a fixed bound above the expected queuing time computed from the
excess window divided by the subpath IP Capacity. This bound must be excess window divided by the subpath IP capacity. This bound must be
smaller than target_RTT/2 to avoid reverting to stop and wait smaller than target_RTT/2 to avoid reverting to stop-and-wait
behavior. (e.g. Data packets and ACKs both have to be released at behavior (e.g., data packets and ACKs both have to be released at
least twice per RTT.) least twice per RTT).
8.3. Slowstart tests 8.3. Slowstart Tests
These tests mimic slowstart: data is sent at twice the effective These tests mimic slowstart: data is sent at twice the effective
bottleneck rate to exercise the queue at the dominant bottleneck. bottleneck rate to exercise the queue at the dominant bottleneck.
8.3.1. Full Window slowstart test 8.3.1. Full Window Slowstart Test
This is a capacity test to confirm that slowstart is not likely to
exit prematurely. Send slowstart bursts that are target_window_size
total packets.
Accumulate packet transfer statistics as described in Section 7.2 to This capacity test confirms that slowstart is not likely to exit
score the outcome. Pass if it is statistically significant that the prematurely. To perform this test, send slowstart bursts that are
observed number of good packets delivered between losses or ECN CE target_window_size total packets and accumulate packet transfer
marks is larger than the target_run_length. Fail if it is statistics as described in Section 7.2 to score the outcome. The
test will pass if it is statistically significant that the observed
number of good packets delivered between losses or ECN CE marks is
larger than the target_run_length. The test will fail if it is
statistically significant that the observed interval between losses statistically significant that the observed interval between losses
or ECN CE marks is smaller than the target_run_length. or ECN CE marks is smaller than the target_run_length.
It is deemed inconclusive if the elapsed time to send the data burst The test is deemed inconclusive if the elapsed time to send the data
is not less than half of the time to receive the ACKs. (i.e. It is burst is not less than half of the time to receive the ACKs. (That
acceptable to send data too fast, but sending it slower than twice is, it is acceptable to send data too fast, but sending it slower
the actual bottleneck rate as indicated by the ACKs is deemed than twice the actual bottleneck rate as indicated by the ACKs is
inconclusive). The headway for the slowstart bursts should be the deemed inconclusive). The headway for the slowstart bursts should be
target_RTT. the target_RTT.
Note that these are the same parameters as the Sender Full Window Note that these are the same parameters that are used for the
burst test, except the burst rate is at slowstart rate, rather than Sustained Full-Rate Bursts Test, except the burst rate is at
sender interface rate. slowstart rate rather than sender interface rate.
8.3.2. Slowstart AQM test 8.3.2. Slowstart AQM Test
Do a continuous slowstart (send data continuously at twice the To perform this test, do a continuous slowstart (send data
implied IP bottleneck capacity), until the first loss, stop, allow continuously at twice the implied IP bottleneck capacity) until the
the network to drain and repeat, gathering statistics on how many first loss; stop and allow the network to drain and repeat; gather
packets were delivered before the loss, the pattern of losses, statistics on how many packets were delivered before the loss, the
maximum observed RTT and window size. Justify the results. There is pattern of losses, maximum observed RTT, and window size; and justify
not currently sufficient theory justifying requiring any particular the results. There is not currently sufficient theory to justify
result, however design decisions that affect the outcome of this requiring any particular result; however, design decisions that
tests also affect how the network balances between long and short affect the outcome of this tests also affect how the network balances
flows (the "mice vs elephants" problem). The queue sojourn time for between long and short flows (the "mice vs. elephants" problem). The
the first packet delivered after the first loss should be at least queue sojourn time for the first packet delivered after the first
one half of the target_RTT. loss should be at least one half of the target_RTT.
This is an engineering test: It should be performed on a quiescent This engineering test should be performed on a quiescent network or
network or testbed, since cross traffic has the potential to change testbed, since cross traffic has the potential to change the results
the results in ill defined ways. in ill-defined ways.
8.4. Sender Rate Burst tests 8.4. Sender Rate Burst Tests
These tests determine how well the network can deliver bursts sent at These tests determine how well the network can deliver bursts sent at
sender's interface rate. Note that this test most heavily exercises the sender's interface rate. Note that this test most heavily
the front path, and is likely to include infrastructure may be out of exercises the front path and is likely to include infrastructure that
scope for an access ISP, even though the bursts might be caused by may be out of scope for an access ISP, even though the bursts might
ACK compression, thinning or channel arbitration in the access ISP. be caused by ACK compression, thinning, or channel arbitration in the
See Appendix B. access ISP. See Appendix B.
Also, there are a several details about sender interface rate bursts Also, there are a several details about sender interface rate bursts
that are not fully defined here. These details, such as the assumed that are not fully defined here. These details, such as the assumed
sender interface rate, should be explicitly stated is a FS-TIDS. sender interface rate, should be explicitly stated in an FSTIDS.
Current standards permit TCP to send full window bursts following an Current standards permit TCP to send full window bursts following an
application pause. (Congestion Window Validation [RFC2861] and application pause. (Congestion Window Validation [RFC2861] and
updates to support Rate-Limited Traffic [RFC7661], are not required). updates to support Rate-Limited Traffic [RFC7661] are not required).
Since full window bursts are consistent with standard behavior, it is Since full window bursts are consistent with standard behavior, it is
desirable that the network be able to deliver such bursts, otherwise desirable that the network be able to deliver such bursts; otherwise,
application pauses will cause unwarranted losses. Note that the AIMD application pauses will cause unwarranted losses. Note that the AIMD
sawtooth requires a peak window that is twice target_window_size, so sawtooth requires a peak window that is twice target_window_size, so
the worst case burst may be 2*target_window_size. the worst-case burst may be 2*target_window_size.
It is also understood in the application and serving community that It is also understood in the application and serving community that
interface rate bursts have a cost to the network that has to be interface rate bursts have a cost to the network that has to be
balanced against other costs in the servers themselves. For example balanced against other costs in the servers themselves. For example,
TCP Segmentation Offload (TSO) reduces server CPU in exchange for TCP Segmentation Offload (TSO) reduces server CPU in exchange for
larger network bursts, which increase the stress on network buffer larger network bursts, which increase the stress on network buffer
memory. Some newer TCP implementations can pace traffic at scale memory. Some newer TCP implementations can pace traffic at scale
[TSO_pacing][TSO_fq_pacing]. It remains to be determined if and how [TSO_pacing] [TSO_fq_pacing]. It remains to be determined if and how
quickly these changes will be deployed. quickly these changes will be deployed.
There is not yet theory to unify these costs or to provide a There is not yet theory to unify these costs or to provide a
framework for trying to optimize global efficiency. We do not yet framework for trying to optimize global efficiency. We do not yet
have a model for how much server rate bursts should be tolerated by have a model for how many server rate bursts should be tolerated by
the network. Some bursts must be tolerated by the network, but it is the network. Some bursts must be tolerated by the network, but it is
probably unreasonable to expect the network to be able to efficiently probably unreasonable to expect the network to be able to efficiently
deliver all data as a series of bursts. deliver all data as a series of bursts.
For this reason, this is the only test for which we encourage For this reason, this is the only test for which we encourage
derating. A TIDS could include a table of pairs of derating derating. A TIDS could include a table containing pairs of derating
parameters: burst sizes and how much each burst size is permitted to parameters: burst sizes and how much each burst size is permitted to
reduce the run length, relative to to the target_run_length. reduce the run length, relative to the target_run_length.
8.5. Combined and Implicit Tests 8.5. Combined and Implicit Tests
Combined tests efficiently confirm multiple network properties in a Combined tests efficiently confirm multiple network properties in a
single test, possibly as a side effect of normal content delivery. single test, possibly as a side effect of normal content delivery.
They require less measurement traffic than other testing strategies They require less measurement traffic than other testing strategies
at the cost of conflating diagnostic signatures when they fail. at the cost of conflating diagnostic signatures when they fail.
These are by far the most efficient for monitoring networks that are These are by far the most efficient for monitoring networks that are
nominally expected to pass all tests. nominally expected to pass all tests.
8.5.1. Sustained Bursts Test 8.5.1. Sustained Full-Rate Bursts Test
The sustained burst test implements a combined worst case version of
all of the capacity tests above. It is simply:
Send target_window_size bursts of packets at server interface rate The Sustained Full-Rate Bursts Test implements a combined worst-case
with target_RTT burst headway (burst start to next burst start). version of all of the capacity tests above. To perform this test,
Verify that the observed packet transfer statistics meets the send target_window_size bursts of packets at server interface rate
with target_RTT burst headway (burst start to next burst start), and
verify that the observed packet transfer statistics meets the
target_run_length. target_run_length.
Key observations: Key observations:
o The subpath under test is expected to go idle for some fraction of o The subpath under test is expected to go idle for some fraction of
the time, determined by the difference between the time to drain the time, determined by the difference between the time to drain
the queue at the subpath_IP_capacity, and the target_RTT. If the the queue at the subpath_IP_capacity and the target_RTT. If the
queue does not drain completely it may be an indication that the queue does not drain completely, it may be an indication that the
the subpath has insufficient IP capacity or that there is some subpath has insufficient IP capacity or that there is some other
other problem with the test (e.g. inconclusive). problem with the test (e.g., it is inconclusive).
o The burst sensitivity can be derated by sending smaller bursts o The burst sensitivity can be derated by sending smaller bursts
more frequently. E.g. send target_window_size*derate packet more frequently (e.g., by sending target_window_size*derate packet
bursts every target_RTT*derate, where "derate" is less than one. bursts every target_RTT*derate, where "derate" is less than one).
o When not derated, this test is the most strenuous capacity test. o When not derated, this test is the most strenuous capacity test.
o A subpath that passes this test is likely to be able to sustain o A subpath that passes this test is likely to be able to sustain
higher rates (close to subpath_IP_capacity) for paths with RTTs higher rates (close to subpath_IP_capacity) for paths with RTTs
significantly smaller than the target_RTT. significantly smaller than the target_RTT.
o This test can be implemented with instrumented TCP [RFC4898], o This test can be implemented with instrumented TCP [RFC4898],
using a specialized measurement application at one end [MBMSource] using a specialized measurement application at one end (e.g.,
and a minimal service at the other end [RFC0863] [RFC0864]. [MBMSource]) and a minimal service at the other end (e.g.,
[RFC863] and [RFC864]).
o This test is efficient to implement, since it does not require o This test is efficient to implement, since it does not require
per-packet timers, and can make use of TSO in modern NIC hardware. per-packet timers, and can make use of TSO in modern network
o If a subpath is known to pass the Standing Queue engineering tests interfaces.
o If a subpath is known to pass the standing queue engineering tests
(particularly that it has a progressive onset of loss at an (particularly that it has a progressive onset of loss at an
appropriate queue depth), then the Sustained Burst Test is appropriate queue depth), then the Sustained Full-Rate Bursts Test
sufficient to assure that the subpath under test will not impair is sufficient to assure that the subpath under test will not
Bulk Transport Capacity at the target performance under all impair Bulk Transport Capacity at the target performance under all
conditions. See Section 8.2 for a discussion of the standing conditions. See Section 8.2 for a discussion of the standing
queue tests. queue tests.
Note that this test is clearly independent of the subpath RTT, or Note that this test is clearly independent of the subpath RTT or
other details of the measurement infrastructure, as long as the other details of the measurement infrastructure, as long as the
measurement infrastructure can accurately and reliably deliver the measurement infrastructure can accurately and reliably deliver the
required bursts to the subpath under test. required bursts to the subpath under test.
8.5.2. Passive Measurements 8.5.2. Passive Measurements
Any non-throughput maximizing application, such as fixed rate Any non-throughput-maximizing application, such as fixed-rate
streaming media, can be used to implement passive or hybrid (defined streaming media, can be used to implement passive or hybrid (defined
in [RFC7799]) versions of Model Based Metrics with some additional in [RFC7799]) versions of Model-Based Metrics with some additional
instrumentation and possibly a traffic shaper or other controls in instrumentation and possibly a traffic shaper or other controls in
the servers. The essential requirement is that the data transmission the servers. The essential requirement is that the data transmission
be constrained such that even with arbitrary application pauses and be constrained such that even with arbitrary application pauses and
bursts, the data rate and burst sizes stay within the envelope bursts, the data rate and burst sizes stay within the envelope
defined by the individual tests described above. defined by the individual tests described above.
If the application's serving data rate can be constrained to be less If the application's serving data rate can be constrained to be less
than or equal to the target_data_rate and the serving_RTT (the RTT than or equal to the target_data_rate and the serving_RTT (the RTT
between the sender and client) is less than the target_RTT, this between the sender and client) is less than the target_RTT, this
constraint is most easily implemented by clamping the transport constraint is most easily implemented by clamping the transport
window size to serving_window_clamp, set to the test_window, computed window size to serving_window_clamp (which is set to the test_window
for the actual serving path. and computed for the actual serving path).
Under the above constraints the serving_window_clamp will limit the Under the above constraints, the serving_window_clamp will limit both
both the serving data rate and burst sizes to be no larger than the the serving data rate and burst sizes to be no larger than the
procedures in Section 8.1.2 and Section 8.4 or Section 8.5.1. Since parameters specified by the procedures in Section 8.1.2, 8.4, or
the serving RTT is smaller than the target_RTT, the worst case bursts 8.5.1. Since the serving RTT is smaller than the target_RTT, the
that might be generated under these conditions will be smaller than worst-case bursts that might be generated under these conditions will
called for by Section 8.4 and the sender rate burst sizes are be smaller than called for by Section 8.4, and the sender rate burst
implicitly derated by the serving_window_clamp divided by the sizes are implicitly derated by the serving_window_clamp divided by
target_window_size at the very least. (Depending on the application the target_window_size at the very least. (Depending on the
behavior, the data might be significantly smoother than specified by application behavior, the data might be significantly smoother than
any of the burst tests.) specified by any of the burst tests.)
In an alternative implementation the data rate and bursts might be In an alternative implementation, the data rate and bursts might be
explicitly controlled by a programmable traffic shaper or pacing at explicitly controlled by a programmable traffic shaper or by pacing
the sender. This would provide better control over transmissions but at the sender. This would provide better control over transmissions
is more complicated to implement, although the required technology is but is more complicated to implement, although the required
available [TSO_pacing][TSO_fq_pacing]. technology is available [TSO_pacing] [TSO_fq_pacing].
Note that these techniques can be applied to any content delivery Note that these techniques can be applied to any content delivery
that can operated at a constrained data rate to inhibit TCP that can be operated at a constrained data rate to inhibit TCP
equilibrium behavior. equilibrium behavior.
Furthermore note that Dynamic Adaptive Streaming over HTTP (DASH) is Furthermore, note that Dynamic Adaptive Streaming over HTTP (DASH) is
generally in conflict with passive Model Based Metrics measurement, generally in conflict with passive Model-Based Metrics measurement,
because it is a rate maximizing protocol. It can still meet the because it is a rate-maximizing protocol. It can still meet the
requirement here if the rate can be capped, for example by knowing a requirement here if the rate can be capped, for example, by knowing a
priori the maximum rate needed to deliver a particular piece of priori the maximum rate needed to deliver a particular piece of
content. content.
9. An Example 9. Example
In this section we illustrate a TIDS designed to confirm that an In this section, we illustrate a TIDS designed to confirm that an
access ISP can reliably deliver HD video from multiple content access ISP can reliably deliver HD video from multiple content
providers to all of their customers. With modern codecs, minimal HD providers to all of its customers. With modern codecs, minimal HD
video (720p) generally fits in 2.5 Mb/s. Due to their geographical video (720p) generally fits in 2.5 Mb/s. Due to the ISP's
size, network topology and modem characteristics the ISP determines geographical size, network topology, and modem characteristics, the
that most content is within a 50 mS RTT of their users (This example ISP determines that most content is within a 50 ms RTT of its users.
RTT is a sufficient to cover the propagation delay to continental (This example RTT is sufficient to cover the propagation delay to
Europe or either US coast with low delay modems or somewhat smaller continental Europe or to either coast of the United States with low-
geographical regions if the modems require additional delay to delay modems; it is sufficient to cover somewhat smaller geographical
implement advanced compression and error recovery). regions if the modems require additional delay to implement advanced
compression and error recovery.)
2.5 Mb/s over a 50 ms path
+----------------------+-------+---------+ +----------------------+-------+---------+
| End-to-End Parameter | value | units | | End-to-End Parameter | value | units |
+----------------------+-------+---------+ +----------------------+-------+---------+
| target_rate | 2.5 | Mb/s | | target_rate | 2.5 | Mb/s |
| target_RTT | 50 | ms | | target_RTT | 50 | ms |
| target_MTU | 1500 | bytes | | target_MTU | 1500 | bytes |
| header_overhead | 64 | bytes | | header_overhead | 64 | bytes |
| | | | | | | |
| target_window_size | 11 | packets | | target_window_size | 11 | packets |
| target_run_length | 363 | packets | | target_run_length | 363 | packets |
+----------------------+-------+---------+ +----------------------+-------+---------+
Table 1 Table 1: 2.5 Mb/s over a 50 ms Path
Table 1 shows the default TCP model with no derating, and as such is Table 1 shows the default TCP model with no derating and, as such, is
quite conservative. The simplest TIDS would be to use the sustained quite conservative. The simplest TIDS would be to use the Sustained
burst test, described in Section 8.5.1. Such a test would send 11 Full-Rate Bursts Test, described in Section 8.5.1. Such a test would
packet bursts every 50mS, and confirming that there was no more than send 11 packet bursts every 50 ms and confirm that there was no more
1 packet loss per 33 bursts (363 total packets in 1.650 seconds). than 1 packet loss per 33 bursts (363 total packets in 1.650
seconds).
Since this number represents is the entire end-to-end loss budget, Since this number represents the entire end-to-end loss budget,
independent subpath tests could be implemented by apportioning the independent subpath tests could be implemented by apportioning the
packet loss ratio across subpaths. For example 50% of the losses packet loss ratio across subpaths. For example, 50% of the losses
might be allocated to the access or last mile link to the user, 40% might be allocated to the access or last mile link to the user, 40%
to the network interconnections with other ISPs and 1% to each to the network interconnections with other ISPs, and 1% to each
internal hop (assuming no more than 10 internal hops). Then all of internal hop (assuming no more than 10 internal hops). Then, all of
the subpaths can be tested independently, and the spatial composition the subpaths can be tested independently, and the spatial composition
of passing subpaths would be expected to be within the end-to-end of passing subpaths would be expected to be within the end-to-end
loss budget. loss budget.
9.1. Observations about applicability 9.1. Observations about Applicability
Guidance on deploying and using MBM belong in a future document. Guidance on deploying and using MBM belong in a future document.
However this example illustrates some the issues that may need to be However, the example above illustrates some of the issues that may
considered. need to be considered.
Note that another ISP, with different geographical coverage, topology Note that another ISP, with different geographical coverage,
or modem technology may need to assume a different target_RTT, and as topology, or modem technology may need to assume a different
a consequence different target_window_size and target_run_length, target_RTT and, as a consequence, a different target_window_size and
even for the same target_data rate. One of the implications of this target_run_length, even for the same target_data rate. One of the
is that infrastructure shared by multiple ISPs, such as inter- implications of this is that infrastructure shared by multiple ISPs,
exchange points (IXPs) and other interconnects may need to be such as Internet Exchange Points (IXPs) and other interconnects may
evaluated on the basis of the most stringent target_window_size and need to be evaluated on the basis of the most stringent
target_run_length of any participating ISP. One way to do this might target_window_size and target_run_length of any participating ISP.
be to choose target parameters for evaluating such shared One way to do this might be to choose target parameters for
infrastructure on the basis of a hypothetical reference path that evaluating such shared infrastructure on the basis of a hypothetical
does not necessarily match any actual paths. reference path that does not necessarily match any actual paths.
Testing interconnects has generally been problematic: conventional Testing interconnects has generally been problematic: conventional
performance tests run between measurement points adjacent to either performance tests run between measurement points adjacent to either
side of the interconnect are not generally useful. Unconstrained TCP side of the interconnect are not generally useful. Unconstrained TCP
tests, such as iPerf [iPerf] are usually overly aggressive due to the tests, such as iPerf [iPerf], are usually overly aggressive due to
small RTT (often less than 1 mS). With a short RTT these tools are the small RTT (often less than 1 ms). With a short RTT, these tools
likely to report inflated data rates because on a short RTT these are likely to report inflated data rates because on a short RTT,
tools can tolerate very high packet loss ratios and can push other these tools can tolerate very high packet loss ratios and can push
cross traffic off of the network. As a consequence these other cross traffic off of the network. As a consequence, these
measurements are useless for predicting actual user performance over measurements are useless for predicting actual user performance over
longer paths, and may themselves be quite disruptive. Model Based longer paths and may themselves be quite disruptive. Model-Based
Metrics solves this problem. The interconnect can be evaluated with Metrics solves this problem. The interconnect can be evaluated with
the same TIDS as other subpaths. Continuing our example, if the the same TIDS as other subpaths. Continuing our example, if the
interconnect is apportioned 40% of the losses, 11 packet bursts sent interconnect is apportioned 40% of the losses, 11 packet bursts sent
every 50mS should have fewer than one loss per 82 bursts (902 every 50 ms should have fewer than one loss per 82 bursts (902
packets). packets).
10. Validation 10. Validation
Since some aspects of the models are likely to be too conservative, Since some aspects of the models are likely to be too conservative,
Section 5.2 permits alternate protocol models and Section 5.3 permits Section 5.2 permits alternate protocol models, and Section 5.3
test parameter derating. If either of these techniques are used, we permits test parameter derating. If either of these techniques is
require demonstrations that such a TIDS can robustly detect subpaths used, we require demonstrations that such a TIDS can robustly detect
that will prevent authentic applications using state-of-the-art subpaths that will prevent authentic applications using state-of-the-
protocol implementations from meeting the specified Target Transport art protocol implementations from meeting the specified Target
Performance. This correctness criteria is potentially difficult to Transport Performance. This correctness criteria is potentially
prove, because it implicitly requires validating a TIDS against all difficult to prove, because it implicitly requires validating a TIDS
possible paths and subpaths. The procedures described here are still against all possible paths and subpaths. The procedures described
experimental. here are still experimental.
We suggest two approaches, both of which should be applied: first, We suggest two approaches, both of which should be applied. First,
publish a fully open description of the TIDS, including what publish a fully open description of the TIDS, including what
assumptions were used and and how it was derived, such that the assumptions were used and how it was derived, such that the research
research community can evaluate the design decisions, test them and community can evaluate the design decisions, test them, and comment
comment on their applicability; and second, demonstrate that on their applicability. Second, demonstrate that applications do
applications do meet the Target Transport Performance when running meet the Target Transport Performance when running over a network
over a network testbed which has the tightest possible constraints testbed that has the tightest possible constraints that still allow
that still allow the tests in the TIDS to pass. the tests in the TIDS to pass.
This procedure resembles an epsilon-delta proof in calculus. This procedure resembles an epsilon-delta proof in calculus.
Construct a test network such that all of the individual tests of the Construct a test network such that all of the individual tests of the
TIDS pass by only small (infinitesimal) margins, and demonstrate that TIDS pass by only small (infinitesimal) margins, and demonstrate that
a variety of authentic applications running over real TCP a variety of authentic applications running over real TCP
implementations (or other protocols as appropriate) meets the Target implementations (or other protocols as appropriate) meets the Target
Transport Performance over such a network. The workloads should Transport Performance over such a network. The workloads should
include multiple types of streaming media and transaction oriented include multiple types of streaming media and transaction-oriented
short flows (e.g. synthetic web traffic). short flows (e.g., synthetic web traffic).
For example, for the HD streaming video TIDS described in Section 9, For example, for the HD streaming video TIDS described in Section 9,
the IP capacity should be exactly the header_overhead above 2.5 Mb/s, the IP capacity should be exactly the header_overhead above 2.5 Mb/s,
the per packet random background loss ratio should be 1/363, for a the per packet random background loss ratio should be 1/363 (for a
run length of 363 packets, the bottleneck queue should be 11 packets run length of 363 packets), the bottleneck queue should be 11
and the front path should have just enough buffering to withstand 11 packets, and the front path should have just enough buffering to
packet interface rate bursts. We want every one of the TIDS tests to withstand 11 packet interface rate bursts. We want every one of the
fail if we slightly increase the relevant test parameter, so for TIDS tests to fail if we slightly increase the relevant test
example sending a 12 packet burst should cause excess (possibly parameter, so, for example, sending a 12-packet burst should cause
deterministic) packet drops at the dominant queue at the bottleneck. excess (possibly deterministic) packet drops at the dominant queue at
This network has the tightest possible constraints that can be the bottleneck. This network has the tightest possible constraints
expected to pass the TIDS, yet it should be possible for a real that can be expected to pass the TIDS, yet it should be possible for
application using a stock TCP implementation in the vendor's default a real application using a stock TCP implementation in the vendor's
configuration to attain 2.5 Mb/s over an 50 mS path. default configuration to attain 2.5 Mb/s over a 50 ms path.
The most difficult part of setting up such a testbed is arranging for The most difficult part of setting up such a testbed is arranging for
it to have the tightest possible constraints that still allow it to it to have the tightest possible constraints that still allow it to
pass the individual tests. Two approaches are suggested: pass the individual tests. Two approaches are suggested:
constraining (configuring) the network devices not to use all
available resources (e.g. by limiting available buffer space or data
rate); and pre-loading subpaths with cross traffic. Note that is it
important that a single tightly constrained environment just barely
passes all tests, otherwise there is a chance that TCP can exploit
extra latitude in some parameters (such as data rate) to partially
compensate for constraints in other parameters (queue space, or vice-
versa).
To the extent that a TIDS is used to inform public dialog it should o constraining (configuring) the network devices not to use all
be fully publicly documented, including the details of the tests, available resources (e.g., by limiting available buffer space or
what assumptions were used and how it was derived. All of the data rate)
o pre-loading subpaths with cross traffic
Note that it is important that a single tightly constrained
environment just barely passes all tests; otherwise, there is a
chance that TCP can exploit extra latitude in some parameters (such
as data rate) to partially compensate for constraints in other
parameters (e.g., queue space). This effect is potentially
bidirectional: extra latitude in the queue space tests has the
potential to enable TCP to compensate for insufficient data-rate
headroom.
To the extent that a TIDS is used to inform public dialog, it should
be fully documented publicly, including the details of the tests,
what assumptions were used, and how it was derived. All of the
details of the validation experiment should also be published with details of the validation experiment should also be published with
sufficient detail for the experiments to be replicated by other sufficient detail for the experiments to be replicated by other
researchers. All components should either be open source of fully researchers. All components should be either open source or fully
described proprietary implementations that are available to the described proprietary implementations that are available to the
research community. research community.
11. Security Considerations 11. Security Considerations
Measurement is often used to inform business and policy decisions, Measurement is often used to inform business and policy decisions
and as a consequence is potentially subject to manipulation. Model and, as a consequence, is potentially subject to manipulation.
Based Metrics are expected to be a huge step forward because Model-Based Metrics are expected to be a huge step forward because
equivalent measurements can be performed from multiple vantage equivalent measurements can be performed from multiple vantage
points, such that performance claims can be independently validated points, such that performance claims can be independently validated
by multiple parties. by multiple parties.
Much of the acrimony in the Net Neutrality debate is due to the Much of the acrimony in the Net Neutrality debate is due to the
historical lack of any effective vantage independent tools to historical lack of any effective vantage-independent tools to
characterize network performance. Traditional methods for measuring characterize network performance. Traditional methods for measuring
Bulk Transport Capacity are sensitive to RTT and as a consequence Bulk Transport Capacity are sensitive to RTT and as a consequence
often yield very different results when run local to an ISP or often yield very different results when run local to an ISP or
interconnect and when run over a customer's complete path. Neither interconnect and when run over a customer's complete path. Neither
the ISP nor customer can repeat the others measurements, leading to the ISP nor customer can repeat the other's measurements, leading to
high levels of distrust and acrimony. Model Based Metrics are high levels of distrust and acrimony. Model-Based Metrics are
expected to greatly improve this situation. expected to greatly improve this situation.
Note that in situ measurements sometimes requires sending synthetic Note that in situ measurements sometimes require sending synthetic
measurement traffic between arbitrary locations in the network, and measurement traffic between arbitrary locations in the network and,
as such are potentially attractive platforms for launching DDOS as such, are potentially attractive platforms for launching DDoS
attacks. All active measurement tools and protocols must be designed attacks. All active measurement tools and protocols must be designed
to minimize the opportunities for these misuses. See the discussion to minimize the opportunities for these misuses. See the discussion
in section 7 of [RFC7594]. in Section 7 of [RFC7594].
Some of the tests described in the note are not intended for frequent
network monitoring since they have the potential to cause high
network loads and might adversely affect other traffic.
This document only describes a framework for designing Fully
Specified Targeted IP Diagnostic Suite. Each FS-TIDS must include
its own security section.
12. Acknowledgments
Ganga Maguluri suggested the statistical test for measuring loss
probability in the target run length. Alex Gilgur and Merry Mou for
helping with the statistics.
Meredith Whittaker for improving the clarity of the communications.
Ruediger Geib provided feedback which greatly improved the document. Some of the tests described in this document are not intended for
frequent network monitoring since they have the potential to cause
high network loads and might adversely affect other traffic.
This work was inspired by Measurement Lab: open tools running on an This document only describes a framework for designing a Fully
open platform, using open tools to collect open data. See Specified Targeted IP Diagnostic Suite. Each FSTIDS must include its
http://www.measurementlab.net/ own security section.
13. IANA Considerations 12. IANA Considerations
This document has no actions for IANA. This document has no IANA actions.
14. Informative References 13. Informative References
[RFC0863] Postel, J., "Discard Protocol", STD 21, RFC 863, May 1983. [RFC863] Postel, J., "Discard Protocol", STD 21, RFC 863,
DOI 10.17487/RFC0863, May 1983,
<https://www.rfc-editor.org/info/rfc863>.
[RFC0864] Postel, J., "Character Generator Protocol", STD 22, [RFC864] Postel, J., "Character Generator Protocol", STD 22,
RFC 864, May 1983. RFC 864, DOI 10.17487/RFC0864, May 1983,
<https://www.rfc-editor.org/info/rfc864>.
[RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, [RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis,
"Framework for IP Performance Metrics", RFC 2330, May "Framework for IP Performance Metrics", RFC 2330,
1998. DOI 10.17487/RFC2330, May 1998,
<https://www.rfc-editor.org/info/rfc2330>.
[RFC2861] Handley, M., Padhye, J., and S. Floyd, "TCP Congestion [RFC2861] Handley, M., Padhye, J., and S. Floyd, "TCP Congestion
Window Validation", RFC 2861, June 2000. Window Validation", RFC 2861, DOI 10.17487/RFC2861, June
2000, <https://www.rfc-editor.org/info/rfc2861>.
[RFC3148] Mathis, M. and M. Allman, "A Framework for Defining [RFC3148] Mathis, M. and M. Allman, "A Framework for Defining
Empirical Bulk Transfer Capacity Metrics", RFC 3148, July Empirical Bulk Transfer Capacity Metrics", RFC 3148,
2001. DOI 10.17487/RFC3148, July 2001,
<https://www.rfc-editor.org/info/rfc3148>.
[RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition
of Explicit Congestion Notification (ECN) to IP", of Explicit Congestion Notification (ECN) to IP",
RFC 3168, DOI 10.17487/RFC3168, September 2001, RFC 3168, DOI 10.17487/RFC3168, September 2001,
<http://www.rfc-editor.org/info/rfc3168>. <https://www.rfc-editor.org/info/rfc3168>.
[RFC3465] Allman, M., "TCP Congestion Control with Appropriate Byte [RFC3465] Allman, M., "TCP Congestion Control with Appropriate Byte
Counting (ABC)", RFC 3465, February 2003. Counting (ABC)", RFC 3465, DOI 10.17487/RFC3465, February
2003, <https://www.rfc-editor.org/info/rfc3465>.
[RFC4015] Ludwig, R. and A. Gurtov, "The Eifel Response Algorithm
for TCP", RFC 4015, February 2005.
[RFC4737] Morton, A., Ciavattone, L., Ramachandran, G., Shalunov, [RFC4737] Morton, A., Ciavattone, L., Ramachandran, G., Shalunov,
S., and J. Perser, "Packet Reordering Metrics", RFC 4737, S., and J. Perser, "Packet Reordering Metrics", RFC 4737,
November 2006. DOI 10.17487/RFC4737, November 2006,
<https://www.rfc-editor.org/info/rfc4737>.
[RFC4898] Mathis, M., Heffner, J., and R. Raghunarayan, "TCP [RFC4898] Mathis, M., Heffner, J., and R. Raghunarayan, "TCP
Extended Statistics MIB", RFC 4898, May 2007. Extended Statistics MIB", RFC 4898, DOI 10.17487/RFC4898,
May 2007, <https://www.rfc-editor.org/info/rfc4898>.
[RFC5136] Chimento, P. and J. Ishac, "Defining Network Capacity", [RFC5136] Chimento, P. and J. Ishac, "Defining Network Capacity",
RFC 5136, February 2008. RFC 5136, DOI 10.17487/RFC5136, February 2008,
<https://www.rfc-editor.org/info/rfc5136>.
[RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion [RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion
Control", RFC 5681, September 2009. Control", RFC 5681, DOI 10.17487/RFC5681, September 2009,
<https://www.rfc-editor.org/info/rfc5681>.
[RFC5827] Allman, M., Avrachenkov, K., Ayesta, U., Blanton, J., and [RFC5827] Allman, M., Avrachenkov, K., Ayesta, U., Blanton, J., and
P. Hurtig, "Early Retransmit for TCP and Stream Control P. Hurtig, "Early Retransmit for TCP and Stream Control
Transmission Protocol (SCTP)", RFC 5827, Transmission Protocol (SCTP)", RFC 5827,
DOI 10.17487/RFC5827, May 2010, DOI 10.17487/RFC5827, May 2010,
<http://www.rfc-editor.org/info/rfc5827>. <https://www.rfc-editor.org/info/rfc5827>.
[RFC5835] Morton, A. and S. Van den Berghe, "Framework for Metric [RFC5835] Morton, A., Ed. and S. Van den Berghe, Ed., "Framework for
Composition", RFC 5835, April 2010. Metric Composition", RFC 5835, DOI 10.17487/RFC5835, April
2010, <https://www.rfc-editor.org/info/rfc5835>.
[RFC6049] Morton, A. and E. Stephan, "Spatial Composition of [RFC6049] Morton, A. and E. Stephan, "Spatial Composition of
Metrics", RFC 6049, January 2011. Metrics", RFC 6049, DOI 10.17487/RFC6049, January 2011,
<https://www.rfc-editor.org/info/rfc6049>.
[RFC6576] Geib, R., Ed., Morton, A., Fardid, R., and A. Steinmitz, [RFC6576] Geib, R., Ed., Morton, A., Fardid, R., and A. Steinmitz,
"IP Performance Metrics (IPPM) Standard Advancement "IP Performance Metrics (IPPM) Standard Advancement
Testing", BCP 176, RFC 6576, DOI 10.17487/RFC6576, March Testing", BCP 176, RFC 6576, DOI 10.17487/RFC6576, March
2012, <http://www.rfc-editor.org/info/rfc6576>. 2012, <https://www.rfc-editor.org/info/rfc6576>.
[RFC6673] Morton, A., "Round-Trip Packet Loss Metrics", RFC 6673, [RFC6673] Morton, A., "Round-Trip Packet Loss Metrics", RFC 6673,
August 2012. DOI 10.17487/RFC6673, August 2012,
<https://www.rfc-editor.org/info/rfc6673>.
[RFC6928] Chu, J., Dukkipati, N., Cheng, Y., and M. Mathis, [RFC6928] Chu, J., Dukkipati, N., Cheng, Y., and M. Mathis,
"Increasing TCP's Initial Window", RFC 6928, "Increasing TCP's Initial Window", RFC 6928,
DOI 10.17487/RFC6928, April 2013, DOI 10.17487/RFC6928, April 2013,
<http://www.rfc-editor.org/info/rfc6928>. <https://www.rfc-editor.org/info/rfc6928>.
[RFC7312] Fabini, J. and A. Morton, "Advanced Stream and Sampling [RFC7312] Fabini, J. and A. Morton, "Advanced Stream and Sampling
Framework for IP Performance Metrics (IPPM)", RFC 7312, Framework for IP Performance Metrics (IPPM)", RFC 7312,
August 2014. DOI 10.17487/RFC7312, August 2014,
<https://www.rfc-editor.org/info/rfc7312>.
[RFC7398] Bagnulo, M., Burbridge, T., Crawford, S., Eardley, P., and [RFC7398] Bagnulo, M., Burbridge, T., Crawford, S., Eardley, P., and
A. Morton, "A Reference Path and Measurement Points for A. Morton, "A Reference Path and Measurement Points for
Large-Scale Measurement of Broadband Performance", Large-Scale Measurement of Broadband Performance",
RFC 7398, February 2015. RFC 7398, DOI 10.17487/RFC7398, February 2015,
<https://www.rfc-editor.org/info/rfc7398>.
[RFC7567] Baker, F., Ed. and G. Fairhurst, Ed., "IETF [RFC7567] Baker, F., Ed. and G. Fairhurst, Ed., "IETF
Recommendations Regarding Active Queue Management", Recommendations Regarding Active Queue Management",
BCP 197, RFC 7567, DOI 10.17487/RFC7567, July 2015, BCP 197, RFC 7567, DOI 10.17487/RFC7567, July 2015,
<http://www.rfc-editor.org/info/rfc7567>. <https://www.rfc-editor.org/info/rfc7567>.
[RFC7594] Eardley, P., Morton, A., Bagnulo, M., Burbridge, T., [RFC7594] Eardley, P., Morton, A., Bagnulo, M., Burbridge, T.,
Aitken, P., and A. Akhter, "A Framework for Large-Scale Aitken, P., and A. Akhter, "A Framework for Large-Scale
Measurement of Broadband Performance (LMAP)", RFC 7594, Measurement of Broadband Performance (LMAP)", RFC 7594,
DOI 10.17487/RFC7594, September 2015, DOI 10.17487/RFC7594, September 2015,
<http://www.rfc-editor.org/info/rfc7594>. <https://www.rfc-editor.org/info/rfc7594>.
[RFC7661] Fairhurst, G., Sathiaseelan, A., and R. Secchi, "Updating [RFC7661] Fairhurst, G., Sathiaseelan, A., and R. Secchi, "Updating
TCP to Support Rate-Limited Traffic", RFC 7661, TCP to Support Rate-Limited Traffic", RFC 7661,
DOI 10.17487/RFC7661, October 2015, DOI 10.17487/RFC7661, October 2015,
<http://www.rfc-editor.org/info/rfc7661>. <https://www.rfc-editor.org/info/rfc7661>.
[RFC7680] Almes, G., Kalidindi, S., Zekauskas, M., and A. Morton, [RFC7680] Almes, G., Kalidindi, S., Zekauskas, M., and A. Morton,
Ed., "A One-Way Loss Metric for IP Performance Metrics Ed., "A One-Way Loss Metric for IP Performance Metrics
(IPPM)", STD 82, RFC 7680, DOI 10.17487/RFC7680, January (IPPM)", STD 82, RFC 7680, DOI 10.17487/RFC7680, January
2016, <http://www.rfc-editor.org/info/rfc7680>. 2016, <https://www.rfc-editor.org/info/rfc7680>.
[RFC7799] Morton, A., "Active and Passive Metrics and Methods (with [RFC7799] Morton, A., "Active and Passive Metrics and Methods (with
Hybrid Types In-Between)", RFC 7799, DOI 10.17487/RFC7799, Hybrid Types In-Between)", RFC 7799, DOI 10.17487/RFC7799,
May 2016, <http://www.rfc-editor.org/info/rfc7799>. May 2016, <https://www.rfc-editor.org/info/rfc7799>.
[I-D.ietf-tcpm-rack] [AFD] Pan, R., Breslau, L., Prabhakar, B., and S. Shenker,
Cheng, Y., Cardwell, N., and N. Dukkipati, "RACK: a time- "Approximate fairness through differential dropping", ACM
based fast loss detection algorithm for TCP", draft-ietf- SIGCOMM Computer Communication Review, Volume 33, Issue 2,
tcpm-rack-02 (work in progress), March 2017. DOI 10.1145/956981.956985, April 2003.
[MSMO97] Mathis, M., Semke, J., Mahdavi, J., and T. Ott, "The [CCscaling]
Macroscopic Behavior of the TCP Congestion Avoidance Paganini, F., Doyle, J., and S. Low, "Scalable laws for
Algorithm", Computer Communications Review volume 27, stable network congestion control", Proceedings of IEEE
number3, July 1997. Conference on Decision and Control,,
DOI 10.1109/CDC.2001.980095, December 2001.
[WPING] Mathis, M., "Windowed Ping: An IP Level Performance [CVST] Krueger, T. and M. Braun, "R package: Fast Cross-
Diagnostic", INET 94, June 1994. Validation via Sequential Testing", version 0.1, 11 2012.
[mpingSource] [iPerf] Wikipedia, "iPerf", November 2017,
Fan, X., Mathis, M., and D. Hamon, "Git Repository for <https://en.wikipedia.org/w/
mping: An IP Level Performance Diagnostic", Sept 2013, index.php?title=Iperf&oldid=810583885>.
<https://github.com/m-lab/mping>.
[MBMSource] [MBMSource]
Hamon, D., Stuart, S., and H. Chen, "Git Repository for "mbm", July 2016, <https://github.com/m-lab/MBM>.
Model Based Metrics", Sept 2013, <https://github.com/m-
lab/MBM>.
[Pathdiag]
Mathis, M., Heffner, J., O'Neil, P., and P. Siemsen,
"Pathdiag: Automated TCP Diagnosis", Passive and Active
Measurement , June 2008.
[iPerf] Wikipedia Contributors, , "iPerf", Wikipedia, The Free
Encyclopedia , cited March 2015,
<http://en.wikipedia.org/w/
index.php?title=Iperf&oldid=649720021>.
[Wald45] Wald, A., "Sequential Tests of Statistical Hypotheses",
The Annals of Mathematical Statistics, Vol. 16, No. 2, pp.
117-186, Published by: Institute of Mathematical
Statistics, Stable URL:
http://www.jstor.org/stable/2235829, June 1945.
[Montgomery90] [Montgomery90]
Montgomery, D., "Introduction to Statistical Quality Montgomery, D., "Introduction to Statistical Quality
Control - 2nd ed.", ISBN 0-471-51988-X, 1990. Control", 2nd Edition, ISBN 0-471-51988-X, 1990.
[Rtool] R Development Core Team, , "R: A language and environment [mpingSource]
for statistical computing. R Foundation for Statistical "mping", July 2016, <https://github.com/m-lab/mping>.
Computing, Vienna, Austria. ISBN 3-900051-07-0, URL
http://www.R-project.org/", , 2011.
[CVST] Krueger, T. and M. Braun, "R package: Fast Cross- [MSMO97] Mathis, M., Semke, J., Mahdavi, J., and T. Ott, "The
Validation via Sequential Testing", version 0.1, 11 2012. Macroscopic Behavior of the TCP Congestion Avoidance
Algorithm", Computer Communications Review, Volume 27,
Issue 3, DOI 10.1145/263932.264023, July 1997.
[AFD] Pan, R., Breslau, L., Prabhakar, B., and S. Shenker, [Pathdiag] Mathis, M., Heffner, J., O'Neil, P., and P. Siemsen,
"Approximate fairness through differential dropping", "Pathdiag: Automated TCP Diagnosis", Passive and Active
SIGCOMM Comput. Commun. Rev. 33, 2, April 2003. Network Measurement, Lecture Notes in Computer Science,
Volume 4979, DOI 10.1007/978-3-540-79232-1_16, 2008.
[wikiBloat] [Policing] Flach, T., Papageorge, P., Terzis, A., Pedrosa, L., Cheng,
Wikipedia, , "Bufferbloat", http://en.wikipedia.org/ Y., Karim, T., Katz-Bassett, E., and R. Govindan, "An
w/ index.php?title=Bufferbloat&oldid=608805474, March Internet-Wide Analysis of Traffic Policing", Proceedings
2015. of ACM SIGCOMM, DOI 10.1145/2934872.2934873, August 2016.
[CCscaling] [RACK] Cheng, Y., Cardwell, N., Dukkipati, N., and P. Jha, "RACK:
Fernando, F., Doyle, J., and S. Steven, "Scalable laws for a time-based fast loss detection algorithm for TCP", Work
stable network congestion control", Proceedings of in Progress, draft-ietf-tcpm-rack-03, March 2018.
Conference on Decision and
Control, http://www.ee.ucla.edu/~paganini, December 2001.
[TSO_pacing] [Rtool] R Development Core Team, "R: A language and environment
Corbet, J., "TSO sizing and the FQ scheduler", for statistical computing", R Foundation for Statistical
LWN.net https://lwn.net/Articles/564978/, Aug 2013. Computing, Vienna, Austria, ISBN 3-900051-07-0, 2011,
<http://www.R-project.org/>.
[TSO_fq_pacing] [TSO_fq_pacing]
Dumazet, E. and Y. Chen, "TSO, fair queuing, pacing: Dumazet, E. and Y. Chen, "TSO, fair queuing, pacing:
three's a charm", Proceedings of IETF 88, TCPM WG three's a charm", Proceedings of IETF 88, TCPM WG,
https://www.ietf.org/proceedings/88/slides/slides-88-tcpm- November 2013,
9.pdf, Nov 2013. <https://www.ietf.org/proceedings/88/slides/
slides-88-tcpm-9.pdf>.
[Policing] [TSO_pacing]
Flach, T., Papageorge, P., Terzis, A., Pedrosa, L., Cheng, Corbet, J., "TSO sizing and the FQ scheduler", August
Y., Karim, T., Katz-Bassett, E., and R. Govindan, "An 2013, <https://lwn.net/Articles/564978/>.
Internet-Wide Analysis of Traffic Policing", ACM SIGCOMM ,
August 2016. [Wald45] Wald, A., "Sequential Tests of Statistical Hypotheses",
The Annals of Mathematical Statistics, Volume 16, Number
2, pp. 117-186, June 1945,
<http://www.jstor.org/stable/2235829>.
[wikiBloat]
Wikipedia, "Bufferbloat", January 2018,
<https://en.wikipedia.org/w/
index.php?title=Bufferbloat&oldid=819293377>.
[WPING] Mathis, M., "Windowed Ping: An IP Level Performance
Diagnostic", Computer Networks and ISDN Systems, Volume
27, Issue 3, DOI 10.1016/0169-7552(94)90119-8, June 1994.
Appendix A. Model Derivations Appendix A. Model Derivations
The reference target_run_length described in Section 5.2 is based on The reference target_run_length described in Section 5.2 is based on
very conservative assumptions: that all excess data in flight very conservative assumptions: that all excess data in flight (i.e.,
(window) above the target_window_size contributes to a standing queue the window size) above the target_window_size contributes to a
that raises the RTT, and that classic Reno congestion control with standing queue that raises the RTT and that classic Reno congestion
delayed ACKs are in effect. In this section we provide two control with delayed ACKs is in effect. In this section we provide
alternative calculations using different assumptions. two alternative calculations using different assumptions.
It may seem out of place to allow such latitude in a measurement It may seem out of place to allow such latitude in a measurement
method, but this section provides offsetting requirements. method, but this section provides offsetting requirements.
The estimates provided by these models make the most sense if network The estimates provided by these models make the most sense if network
performance is viewed logarithmically. In the operational Internet, performance is viewed logarithmically. In the operational Internet,
data rates span more than 8 orders of magnitude, RTT spans more than data rates span more than eight orders of magnitude, RTT spans more
3 orders of magnitude, and packet loss ratio spans at least 8 orders than three orders of magnitude, and packet loss ratio spans at least
of magnitude if not more. When viewed logarithmically (as in eight orders of magnitude if not more. When viewed logarithmically
decibels), these correspond to 80 dB of dynamic range. On an 80 dB (as in decibels), these correspond to 80 dB of dynamic range. On an
scale, a 3 dB error is less than 4% of the scale, even though it 80 dB scale, a 3 dB error is less than 4% of the scale, even though
represents a factor of 2 in untransformed parameter. it represents a factor of 2 in untransformed parameter.
This document gives a lot of latitude for calculating This document gives a lot of latitude for calculating
target_run_length, however people designing a TIDS should consider target_run_length; however, people designing a TIDS should consider
the effect of their choices on the ongoing tussle about the relevance the effect of their choices on the ongoing tussle about the relevance
of "TCP friendliness" as an appropriate model for Internet capacity of "TCP friendliness" as an appropriate model for Internet capacity
allocation. Choosing a target_run_length that is substantially allocation. Choosing a target_run_length that is substantially
smaller than the reference target_run_length specified in Section 5.2 smaller than the reference target_run_length specified in Section 5.2
strengthens the argument that it may be appropriate to abandon "TCP strengthens the argument that it may be appropriate to abandon "TCP
friendliness" as the Internet fairness model. This gives developers friendliness" as the Internet fairness model. This gives developers
incentive and permission to develop even more aggressive applications incentive and permission to develop even more aggressive applications
and protocols, for example by increasing the number of connections and protocols, for example, by increasing the number of connections
that they open concurrently. that they open concurrently.
A.1. Queueless Reno A.1. Queueless Reno
In Section 5.2 models were derived based on the assumption that the In Section 5.2, models were derived based on the assumption that the
subpath IP rate matches the target rate plus overhead, such that the subpath IP rate matches the target rate plus overhead, such that the
excess window needed for the AIMD sawtooth causes a fluctuating queue excess window needed for the AIMD sawtooth causes a fluctuating queue
at the bottleneck. at the bottleneck.
An alternate situation would be a bottleneck where there is no An alternate situation would be a bottleneck where there is no
significant queue and losses are caused by some mechanism that does significant queue and losses are caused by some mechanism that does
not involve extra delay, for example by the use of a virtual queue as not involve extra delay, for example, by the use of a virtual queue
done in Approximate Fair Dropping [AFD]. A flow controlled by such a as done in Approximate Fair Dropping [AFD]. A flow controlled by
bottleneck would have a constant RTT and a data rate that fluctuates such a bottleneck would have a constant RTT and a data rate that
in a sawtooth due to AIMD congestion control. Assume the losses are fluctuates in a sawtooth due to AIMD congestion control. Assume the
being controlled to make the average data rate meet some goal which losses are being controlled to make the average data rate meet some
is equal or greater than the target_rate. The necessary run length goal that is equal to or greater than the target_rate. The necessary
to meet the target_rate can be computed as follows: run length to meet the target_rate can be computed as follows:
For some value of Wmin, the window will sweep from Wmin packets to For some value of Wmin, the window will sweep from Wmin packets to
2*Wmin packets in 2*Wmin RTT (due to delayed ACK). Unlike the 2*Wmin packets in 2*Wmin RTT (due to delayed ACK). Unlike the
queuing case where Wmin = target_window_size, we want the average of queuing case where Wmin = target_window_size, we want the average of
Wmin and 2*Wmin to be the target_window_size, so the average data Wmin and 2*Wmin to be the target_window_size, so the average data
rate is the target rate. Thus we want Wmin = rate is the target rate. Thus, we want Wmin =
(2/3)*target_window_size. (2/3)*target_window_size.
Between losses each sawtooth delivers (1/2)(Wmin+2*Wmin)(2Wmin) Between losses, each sawtooth delivers (1/2)(Wmin+2*Wmin)(2Wmin)
packets in 2*Wmin round trip times. packets in 2*Wmin RTTs.
Substituting these together we get: Substituting these together, we get:
target_run_length = (4/3)(target_window_size^2) target_run_length = (4/3)(target_window_size^2)
Note that this is 44% of the reference_run_length computed earlier. Note that this is 44% of the reference_run_length computed earlier.
This makes sense because under the assumptions in Section 5.2 the This makes sense because under the assumptions in Section 5.2, the
AMID sawtooth caused a queue at the bottleneck, which raised the AMID sawtooth caused a queue at the bottleneck, which raised the
effective RTT by 50%. effective RTT by 50%.
Appendix B. The effects of ACK scheduling Appendix B. The Effects of ACK Scheduling
For many network technologies simple queuing models don't apply: the For many network technologies, simple queuing models don't apply: the
network schedules, thins or otherwise alters the timing of ACKs and network schedules, thins, or otherwise alters the timing of ACKs and
data, generally to raise the efficiency of the channel allocation data, generally to raise the efficiency of the channel allocation
algorithms when confronted with relatively widely spaced small ACKs. algorithms when confronted with relatively widely spaced small ACKs.
These efficiency strategies are ubiquitous for half duplex, wireless These efficiency strategies are ubiquitous for half-duplex, wireless,
and broadcast media. and broadcast media.
Altering the ACK stream by holding or thinning ACKs typically has two Altering the ACK stream by holding or thinning ACKs typically has two
consequences: it raises the implied bottleneck IP capacity, making consequences: it raises the implied bottleneck IP capacity, making
the fine grained slowstart bursts either faster or larger and it the fine-grained slowstart bursts either faster or larger, and it
raises the effective RTT by the average time that the ACKs and data raises the effective RTT by the average time that the ACKs and data
are delayed. The first effect can be partially mitigated by re- are delayed. The first effect can be partially mitigated by
clocking ACKs once they are beyond the bottleneck on the return path re-clocking ACKs once they are beyond the bottleneck on the return
to the sender, however this further raises the effective RTT. path to the sender; however, this further raises the effective RTT.
The most extreme example of this sort of behavior would be a half The most extreme example of this sort of behavior would be a half-
duplex channel that is not released as long as the endpoint currently duplex channel that is not released as long as the endpoint currently
holding the channel has more traffic (data or ACKs) to send. Such holding the channel has more traffic (data or ACKs) to send. Such
environments cause self clocked protocols under full load to revert environments cause self-clocked protocols under full load to revert
to extremely inefficient stop and wait behavior. The channel to extremely inefficient stop-and-wait behavior. The channel
constrains the protocol to send an entire window of data as a single constrains the protocol to send an entire window of data as a single
contiguous burst on the forward path, followed by the entire window contiguous burst on the forward path, followed by the entire window
of ACKs on the return path. of ACKs on the return path. (A channel with this behavior would fail
the Duplex Self-Interference Test described in Section 8.2.4).
If a particular return path contains a subpath or device that alters If a particular return path contains a subpath or device that alters
the timing of the ACK stream, then the entire front path from the the timing of the ACK stream, then the entire front path from the
sender up to the bottleneck must be tested at the burst parameters sender up to the bottleneck must be tested at the burst parameters
implied by the ACK scheduling algorithm. The most important implied by the ACK scheduling algorithm. The most important
parameter is the Implied Bottleneck IP Capacity, which is the average parameter is the implied bottleneck IP capacity, which is the average
rate at which the ACKs advance snd.una. Note that thinning the ACK rate at which the ACKs advance snd.una. Note that thinning the ACK
stream (relying on the cumulative nature of seg.ack to permit stream (relying on the cumulative nature of seg.ack to permit
discarding some ACKs) causes most TCP implementations to send discarding some ACKs) causes most TCP implementations to send
interface rate bursts to offset the longer times between ACKs in interface rate bursts to offset the longer times between ACKs in
order to maintain the average data rate. order to maintain the average data rate.
Note that due to ubiquitous self clocking in Internet protocols, ill Note that due to ubiquitous self-clocking in Internet protocols,
conceived channel allocation mechanisms are likely to increases the ill-conceived channel allocation mechanisms are likely to increases
queuing stress on the front path because they cause larger full the queuing stress on the front path because they cause larger full
sender rate data bursts. sender rate data bursts.
Holding data or ACKs for channel allocation or other reasons (such as Holding data or ACKs for channel allocation or other reasons (such as
forward error correction) always raises the effective RTT relative to forward error correction) always raises the effective RTT relative to
the minimum delay for the path. Therefore it may be necessary to the minimum delay for the path. Therefore, it may be necessary to
replace target_RTT in the calculation in Section 5.2 by an replace target_RTT in the calculation in Section 5.2 by an
effective_RTT, which includes the target_RTT plus a term to account effective_RTT, which includes the target_RTT plus a term to account
for the extra delays introduced by these mechanisms. for the extra delays introduced by these mechanisms.
Appendix C. Version Control Acknowledgments
This section to be removed prior to publication. Ganga Maguluri suggested the statistical test for measuring loss
probability in the target run length. Alex Gilgur and Merry Mou
helped with the statistics.
Formatted: Thu Apr 7 18:12:37 PDT 2016 Meredith Whittaker improved the clarity of the communications.
Ruediger Geib provided feedback that greatly improved the document.
This work was inspired by Measurement Lab: open tools running on an
open platform, using open tools to collect open data. See
<http://www.measurementlab.net/>.
Authors' Addresses Authors' Addresses
Matt Mathis Matt Mathis
Google, Inc Google, Inc
1600 Amphitheater Parkway 1600 Amphitheatre Parkway
Mountain View, California 94043 Mountain View, CA 94043
USA United States of America
Email: mattmathis@google.com Email: mattmathis@google.com
Al Morton Al Morton
AT&T Labs AT&T Labs
200 Laurel Avenue South 200 Laurel Avenue South
Middletown, NJ 07748 Middletown, NJ 07748
USA United States of America
Phone: +1 732 420 1571 Phone: +1 732 420 1571
Email: acmorton@att.com Email: acmorton@att.com
 End of changes. 422 change blocks. 
1394 lines changed or deleted 1349 lines changed or added

This html diff was produced by rfcdiff 1.46. The latest version is available from http://tools.ietf.org/tools/rfcdiff/