draft-ietf-nvo3-encap-05.txt   draft-ietf-nvo3-encap-06.txt 
NVO3 Workgroup S. Boutros, Ed. NVO3 Workgroup S. Boutros, Ed.
Internet-Draft Ciena Internet-Draft Ciena
Intended status: Informational February 17, 2020 Intended Status: Informational D. Eastlake, Ed.
Expires: August 20, 2020 Futurewei
Expires: December 8, 2021 June 9, 2021
NVO3 Encapsulation Considerations NVO3 Encapsulation Considerations
draft-ietf-nvo3-encap-05 draft-ietf-nvo3-encap-06
Abstract Abstract
As communicated by the WG Chairs, the IETF NVO3 chairs and Routing
As communicated by WG Chairs, the IETF NVO3 chairs and Routing Area Area director have chartered a design team to take forward the
director have chartered a design team to take forward the
encapsulation discussion and see if there is potential to design a encapsulation discussion and see if there is potential to design a
common encapsulation that addresses the various technical concerns. common encapsulation that addresses the various technical concerns.
There are implications of different encapsulations in real There are implications of different encapsulations in real
environments consisting of both software and hardware implementations environments consisting of both software and hardware implementations
and spanning multiple data centers. For example, OAM functions such and spanning multiple data centers. For example, OAM functions such
as path MTU discovery become challenging with multiple encapsulations as path MTU discovery become challenging with multiple encapsulations
along the data path. along the data path.
The design team recommend Geneve with few modifications as the common The design team recommends Geneve with a few modifications as the
encapsulation, more details are described in section 7. common encapsulation. This document provides more details,
particularly in Section 7.
Status of This Memo Status of This Document
This Internet-Draft is submitted in full conformance with the This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79. provisions of BCP 78 and BCP 79.
Distribution of this document is unlimited. Comments should be sent
to the authors or the IDR Working Group mailing list <nvo3@ietf.org>.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute Task Force (IETF), its areas, and its working groups. Note that
working documents as Internet-Drafts. The list of current Internet- other groups may also distribute working documents as Internet-
Drafts is at https://datatracker.ietf.org/drafts/current/. Drafts.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
This Internet-Draft will expire on August 20, 2020. The list of current Internet-Drafts can be accessed at
https://www.ietf.org/1id-abstracts.html. The list of Internet-Draft
Shadow Directories can be accessed at
https://www.ietf.org/shadow.html.
Copyright Notice Copyright Notice
Internet-Draft NVO3 Encapsulation Considerations
Copyright (c) 2020 IETF Trust and the persons identified as the Copyright (c) 2021 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
(https://trustee.ietf.org/license-info) in effect on the date of (http://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.
Internet-Draft NVO3 Encapsulation Considerations
Table of Contents Table of Contents
1. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 3 1. Introduction............................................4
2. Design Team Goals . . . . . . . . . . . . . . . . . . . . . . 3 2. Design Team Goals.......................................4
3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 3. Terminology.............................................5
4. Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . 4 4. Abbreviations and Acronyms..............................5
5. Issues with current Encapsulations . . . . . . . . . . . . . 4
5.1. Geneve . . . . . . . . . . . . . . . . . . . . . . . . . 4
5.2. GUE . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
5.3. VXLAN-GPE . . . . . . . . . . . . . . . . . . . . . . . . 5
6. Common Encapsulation Considerations . . . . . . . . . . . . . 5
6.1. Current Encapsulations . . . . . . . . . . . . . . . . . 5
6.2. Useful Extensions Use cases . . . . . . . . . . . . . . . 5
6.2.1. Telemetry extensions. . . . . . . . . . . . . . . . . 6
6.2.2. Security/Integrity extensions . . . . . . . . . . . . 6
6.2.3. Group Base Policy . . . . . . . . . . . . . . . . . . 7
6.3. Hardware Considerations . . . . . . . . . . . . . . . . . 7
6.4. Extension Size . . . . . . . . . . . . . . . . . . . . . 7
6.5. Extension Ordering . . . . . . . . . . . . . . . . . . . 8
6.6. TLV vs Bit Fields . . . . . . . . . . . . . . . . . . . . 8
6.7. Control Plane Considerations . . . . . . . . . . . . . . 9
6.8. Split NVE . . . . . . . . . . . . . . . . . . . . . . . . 10
6.9. Larger VNI Considerations . . . . . . . . . . . . . . . . 11
7. Design team recommendations . . . . . . . . . . . . . . . . . 11
8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 13
9. Security Considerations . . . . . . . . . . . . . . . . . . . 14
10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 14
11. Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . 14
11.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . 14
11.2. Extensibility . . . . . . . . . . . . . . . . . . . . . 14
11.2.1. Native Extensibility Support . . . . . . . . . . . . 14
11.2.2. Extension Parsing . . . . . . . . . . . . . . . . . 14
11.2.3. Critical Extensions . . . . . . . . . . . . . . . . 15
11.2.4. Maximal Header Length . . . . . . . . . . . . . . . 15
11.3. Encapsulation Header . . . . . . . . . . . . . . . . . . 15
11.3.1. Virtual Network Identifier (VNI) . . . . . . . . . . 15
11.3.2. Next Protocol . . . . . . . . . . . . . . . . . . . 15
11.3.3. Other Header Fields . . . . . . . . . . . . . . . . 16
11.4. Comparison Summary . . . . . . . . . . . . . . . . . . . 16 5. Issues with Current Encapsulations......................6
12. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 17 5.1. Geneve................................................6
13. References . . . . . . . . . . . . . . . . . . . . . . . . . 18 5.2. GUE...................................................6
13.1. Normative References . . . . . . . . . . . . . . . . . . 18 5.3. VXLAN-GPE.............................................6
13.2. Informative References . . . . . . . . . . . . . . . . . 18
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 19
1. Problem Statement 6. Common Encapsulation Considerations.....................7
6.1. Current Encapsulations................................7
6.2. Useful Extensions Use Cases...........................7
6.2.1. Telemetry Extensions................................7
6.2.2. Security/Integrity Extensions.......................8
6.2.3 Group Base Policy....................................8
6.3. Hardware Considerations...............................9
6.4. Extension Size........................................9
6.5. Extension Ordering...................................10
6.6. TLV versus Bit Fields................................10
6.7. Control Plane Considerations.........................11
6.8. Split NVE............................................12
6.9. Larger VNI Considerations............................12
As communicated by WG Chairs, the NVO3 WG charter states that it may 7. Design Team Recommendations............................13
produce requirements for network virtualization data planes based on 8. Acknowledgements.......................................16
encapsulation of virtual network traffic over an IP-based underlay
9. Security Considerations................................16
10. IANA Considerations...................................16
11. References............................................17
11.1 Normative References.................................17
11.2 Informative References...............................17
Appendix A: Encapsulations Comparison.....................19
A.1. Overview.............................................19
A.2. Extensibility........................................19
A.2.1. Native Extensibility Support.......................19
A.2.2. Extension Parsing..................................19
A.2.3. Critical Extensions................................20
A.2.4. Maximal Header Length..............................20
A.3. Encapsulation Header.................................20
A.3.1. Virtual Network Identifier (VNI)...................20
A.3.2. Next Protocol......................................20
A.3.3. Other Header Fields................................21
A.4. Comparison Summary...................................21
Contributors..............................................23
Internet-Draft NVO3 Encapsulation Considerations
1. Introduction
As communicated by the WG Chairs, the NVO3 WG Charter states that it
may produce requirements for network virtualization data planes based
on encapsulation of virtual network traffic over an IP-based underlay
data plane. Such requirements should consider OAM and security. data plane. Such requirements should consider OAM and security.
Based on these requirements the WG will select, extend, and/or Based on these requirements the WG will select, extend, and/or
develop one or more data plane encapsulation format(s). develop one or more data plane encapsulation format(s).
This has led to drafts describing three encapsulations being adopted This has led to WG drafts and an RFC describing three encapsulations
by the working group: as follows:
- [I-D.ietf-nvo3-geneve] - [RFC8926] Geneve: Generic Network Virtualization Encapsulation
- [I-D.ietf-nvo3-gue] - [I-D.ietf-intarea-gue] Generic UDP Encapsulation
- [I-D.ietf-nvo3-vxlan-gpe] - [I-D.ietf-nvo3-vxlan-gpe] Generic Protocol Extension for VXLAN
(VXLAN-GPE)
Discussion on the list and in face-to-face meetings has identified a Discussion on the list and in face-to-face meetings has identified a
number of technical problems with each of these encapsulations. number of technical problems with each of these encapsulations.
Furthermore, there was clear consensus at the IETF meeting in Berlin Furthermore, there was clear consensus at the 96th IETF meeting in
that it is undesirable for the working group to progress more than Berlin that it is undesirable for the working group to progress more
one data plane encapsulation. Although consensus could not be than one data plane encapsulation. Although consensus could not be
reached on the list, the overall consensus was for a single reached on the list, the overall consensus was for a single
encapsulation [RFC2418],Section 3.3. encapsulation [RFC2418], Section 3.3.
Nonetheless there has been resistance to converging on a single Nonetheless there has been resistance to converging on a single
encapsulation format. encapsulation format.
2. Design Team Goals 2. Design Team Goals
As communicated by WG Chairs, the design team should take one of the As communicated by the WG Chairs, the design team should take one of
proposed encapsulations and enhance it to address the technical the proposed encapsulations and enhance it to address the technical
concerns. The simple evolution of deployed networks as well as concerns. The simple evolution of deployed networks as well as
applicability to all locations in the NVO3 architecture are goals. applicability to all locations in the NVO3 architecture are goals.
The DT should specifically avoid a design that is burdensome on The DT should specifically avoid a design that is burdensome on
hardware implementations, but should allow future extensibility. The hardware implementations but should allow future extensibility. The
chosen design should also operate well with ICMP and in ECMP chosen design should also operate well with ICMP and in ECMP
environments. If further extensibility is required, then it should environments. If further extensibility is required, then it should
be done in such a manner that it does not require the consent of an be done in such a manner that it does not require the consent of an
entity outside of the IETF. entity outside of the IETF.
3. Terminology Internet-Draft NVO3 Encapsulation Considerations
3. Terminology
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
document are to be interpreted as described in [RFC2119]. "OPTIONAL" in this document are to be interpreted as described in BCP
14 [RFC2119] [RFC8174] when, and only when, they appear in all
capitals, as shown here.
4. Abbreviations 4. Abbreviations and Acronyms
NVO3 Network Virtualization Overlays over Layer 3 DT NVO3 encapsulation Design Team
OAM Operations, Administration, and Maintenance NVO3 Network Virtualization Overlays over Layer 3
TLV Type, Length, and Value OAM Operations, Administration, and Maintenance
VNI Virtual Network Identifier TLV Type, Length, and Value
NVE Network Virtualization Edge VNI Virtual Network Identifier
NVA Network Virtualization Authority NVE Network Virtualization Edge
NIC Network interface card NVA Network Virtualization Authority
Transit device Underlay network devices between NVE(s). NIC Network interface card
5. Issues with current Encapsulations TCAM Ternary Content-Addressable Memory
As summarized by WG Chairs. Transit device - Underlay network devices between NVE(s).
5.1. Geneve Internet-Draft NVO3 Encapsulation Considerations
5. Issues with Current Encapsulations
The following subsections describe issues with current encapsulations
as summarized by the WG Chairs:
5.1. Geneve
- Can't be implemented cost-effectively in all use cases because - Can't be implemented cost-effectively in all use cases because
variable length header and order of the TLVs makes is costly (in variable length header and order of the TLVs makes is costly (in
terms of number of gates) to implement in hardware terms of number of gates) to implement in hardware.
- Header doesn't fit into largest commonly available parse buffer - Header doesn't fit into largest commonly available parse buffer
(256 bytes in NIC). Cannot justify doubling buffer size unless it is (256 bytes in NIC). Cannot justify doubling buffer size unless it is
mandatory for hardware to process additional option fields. mandatory for hardware to process additional option fields.
5.2. GUE 5.2. GUE
- There were a significant number of objections related to the - There were a significant number of objections related to the
complexity of implementation in hardware, similar to those noted for complexity of implementation in hardware, similar to those noted for
Geneve above. Geneve above.
5.3. VXLAN-GPE 5.3. VXLAN-GPE
- GPE is not day-1 backwards compatible with VXLAN. Although the - GPE is not day-1 backwards compatible with VXLAN. Although the
frame format is similar, it uses a different UDP port, so would frame format is similar, it uses a different UDP port, so would
require changes to existing implementations even if the rest of the require changes to existing implementations even if the rest of the
GPE frame is the same. GPE frame is the same.
- GPE is insufficiently extensible. Numerous extensions and options - GPE is insufficiently extensible. Numerous extensions and options
have been designed for GUE and Geneve. Note that these have not yet have been designed for GUE and Geneve. Note that these have not yet
been validated by the WG. been validated by the WG.
- Security e.g. of the VNI has not been addressed by GPE. Although a - Security, e.g., of the VNI, has not been addressed by GPE.
shim header could be used for security and other extensions, this has Although a shim header could be used for security and other
not been defined yet and its implications on offloading in NICs are extensions, this has not been defined yet and its implications on
not understood. offloading in NICs are not understood.
6. Common Encapsulation Considerations Internet-Draft NVO3 Encapsulation Considerations
6.1. Current Encapsulations 6. Common Encapsulation Considerations
6.1. Current Encapsulations
Appendix A includes a detailed comparison between the three proposed Appendix A includes a detailed comparison between the three proposed
encapsulations. The comparison indicates several common properties, encapsulations. The comparison indicates several common properties
but also three major differences among the encapsulations: but also three major differences among the encapsulations:
- Extensibility: Geneve and GUE were defined with built-in - Extensibility: Geneve and GUE were defined with built-in
extensibility, while VXLAN-GPE is not inherently extensible. Note extensibility, while VXLAN-GPE is not inherently extensible. Note
that any of the three encapsulations can be extended using the that any of the three encapsulations can be extended using the
Network Service Header (NSH). Network Service Header (NSH [RFC8300]).
- Extension method: Geneve is extensible using Type/Length/Value - Extension method: Geneve is extensible using Type/Length/Value
(TLV) fields, while GUE uses a small set of possible extensions, and (TLV) fields, while GUE uses a small set of possible extensions, and
a set of flags that indicate which extension is present. a set of flags that indicate which extensions are present.
- Length field: Geneve and GUE include a Length field, indicating the - Length field: Geneve and GUE include a Length field, indicating the
length of the encapsulation header, while VXLAN-GPE does not include length of the encapsulation header while VXLAN-GPE does not include
such a field. such a field.
6.2. Useful Extensions Use cases 6.2. Useful Extensions Use Cases
Non vendor specific TLV MUST follow the standardization process. The Non vendor specific TLVs MUST follow the standardization process.
following use cases for extensions shows that there is a strong The following use cases for extensions shows that there is a strong
requirement to support variable length extensions with possible requirement to support variable length extensions with possible
different subtypes. different subtypes.
6.2.1. Telemetry extensions. 6.2.1. Telemetry Extensions
In several scenarios it is beneficial to make information about the In several scenarios it is beneficial to make information about the
path a packet took through the network or through a network device as path a packet took through the network or through a network device as
well as associated telemetry information available to the operator. well as associated telemetry information available to the operator.
This includes not only tasks like debugging, troubleshooting, as well This includes not only tasks like debugging, troubleshooting, and
as network planning and network optimization but also policy or network planning and optimization but also policy or service level
service level agreement compliance checks. agreement compliance checks.
Packet scheduling algorithms, especially for balancing traffic across Packet scheduling algorithms, especially for balancing traffic across
equal cost paths or links, often leverage information contained equal cost paths or links, often leverage information contained
within the packet, such as protocol number, IP-address or MAC- within the packet, such as protocol number, IP-address, or MAC-
address. Probe packets would thus either need to be sent from the address. Probe packets would thus either need to be sent between the
exact same endpoints with the exact same parameters, or probe packets exact same endpoints with the exact same parameters, or probe packets
would need to be artificially constructed as "fake" packets and would need to be artificially constructed as "fake" packets and
Internet-Draft NVO3 Encapsulation Considerations
inserted along the path. Both approaches are often not feasible from inserted along the path. Both approaches are often not feasible from
an operational perspective, be it that access to the end-system is an operational perspective, be it that access to the end-system is
not feasible, or that the diversity of parameters and associated not feasible, or that the diversity of parameters and associated
probe packets to be created is simply too large. An in-band probe packets to be created is simply too large. An extension
telemetry mechanism in extensions is an alternative in those cases. providing an in-band telemetry mechanism is an alternative in those
cases.
6.2.2. Security/Integrity extensions 6.2.2. Security/Integrity Extensions
Since the currently proposed NVO3 encapsulations do not protect their Since the currently proposed NVO3 encapsulations do not protect their
headers a single bit corruption in the VNI field could deliver a headers, a single bit corruption in the VNI field could deliver a
packet to the wrong tenant. Extensions are needed to use any packet to the wrong tenant. Extensions are needed to use any
sophisticated security. sophisticated security.
The possibility of VNI spoofing with an NVO3 protocol is exacerbated The possibility of VNI spoofing with an NVO3 protocol is exacerbated
by the use of UDP. Systems typically have no restrictions on by using UDP. Systems typically have no restrictions on applications
applications being able to send to any UDP port so an unprivileged being able to send to any UDP port so an unprivileged application can
application can trivially spoof for instance, VXLAN packets, trivially spoof VXLAN packets for instance, including using arbitrary
including using arbitrary VNIs. VNIs.
One can envision HMAC-like support in some NVO3 extension to One can envision HMAC-like support in some NVO3 extension to
authenticate the header and the outer IP addresses, thereby authenticate the header and the outer IP addresses, thereby
preventing attackers from injecting packets with spoofed VNIs. preventing attackers from injecting packets with spoofed VNIs.
An other aspect of security is payload security. Essentially this is Another aspect of security is payload security. Essentially this is
to make packets that look like IP|UDP|NVO3 Encap|DTLS/IPSEC-ESP to make packets that look like IP|UDP|NVO3 Encap|DTLS/IPSEC-ESP
Extension|payload. This is nice since we still have the UDP header Extension|payload. This is nice since we still have the UDP header
for ECMP, the NVO3 header is in plain text so it can by read by for ECMP, the NVO3 header is in plain text so it can be read by
network elements, and different security or other payload transforms network elements, and different security or other payload transforms
can be supported on a single UDP port (we don't need a separate UDP can be supported on a single UDP port (we don't need a separate UDP
for DTLS/IPSEC). for DTLS/IPSEC).
6.2.3. Group Base Policy 6.2.3 Group Base Policy
Another use case would be to carry the Group Based Policy (GBP) Another use case would be to carry the Group Based Policy (GBP)
source group information within a NVO3 header extension in a similar source group information within a NVO3 header extension in a similar
manner as has been implemented for VXLAN manner as has been implemented for VXLAN
[I-D.smith-vxlan-group-policy]. This allows various forms of policy [I-D.smith-vxlan-group-policy]. This allows various forms of policy
such as access control and QoS to be applied between abstract groups such as access control and QoS to be applied between abstract groups
rather than coupled to specific endpoint addresses. rather than coupled to specific endpoint addresses.
6.3. Hardware Considerations Internet-Draft NVO3 Encapsulation Considerations
6.3. Hardware Considerations
Hardware restrictions should be taken into consideration along with Hardware restrictions should be taken into consideration along with
future hardware enhancements that may provide more flexible metadata future hardware enhancements that may provide more flexible metadata
processing. However, the set of options that need to and will be processing. However, the set of options that need to and will be
implemented in hardware will be a subset of what is implemented in implemented in hardware will be a subset of what is implemented in
software, since software NVEs are likely to grow features, and hence software, since software NVEs are likely to grow features, and hence
option support, at a more rapid rate. option support, at a more rapid rate.
We note that it is hard to predict which options will be implemented We note that it is hard to predict which options will be implemented
in which piece of hardware and when. That depends on whether the in which piece of hardware and when. That depends on whether the
hardware will be in the form of a NIC providing increasing offload hardware will be in the form of a NIC providing increasing offload
capabilities to software NVEs, or a switch chip being used as an NVE capabilities to software NVEs, or a switch chip being used as an NVE
gateway towards non-NVO3 parts of the network, or even an transit gateway towards non-NVO3 parts of the network, or even a transit
devices that participates in the NVO3 dataplane e.g. for OAM device that participates in the NVO3 dataplane, e.g., for OAM
purposes. purposes.
A result of this is that it doesn't look useful to prescribe some A result of this is that it doesn't look useful to prescribe some
order of the option so that the ones that are likely to be order of the option so that the ones that are likely to be
implemented in hardware come first; we can't decide such an order implemented in hardware come first; we can't decide such an order
when we define the options, however a control plane can enforce such when we define the options, however a control plane can enforce such
order for some hardware implementations. an order for some hardware implementation.
We do know that hardware needs to initially be able to efficiently We do know that hardware needs to initially be able to efficiently
skip over the NVO3 header to find the inner payload. That is needed skip over the NVO3 header to find the inner payload. That is needed
for both NICs doing e.g. TCP offload and transit devices and NVEs both for NICs doing TCP offload and for transit devices and NVEs
applying policy/ACLs to the inner payload. applying policy/ACLs to the inner payload.
6.4. Extension Size 6.4. Extension Size
Extension header length has a significant impact to hardware and Extension header length has a significant impact on hardware and
software implementations. A total header length that is too small software implementations. A total header length that is too small
will unnecessarily constrained software flexibility. A total header will unnecessarily constrained software flexibility. A total header
length that is too large will place a nontrivial cost on hardware length that is too large will place a nontrivial cost on hardware
implementations. Thus, the design team recommends that there be a implementations. Thus, the design team recommends that there be a
minimum and maximum total extension header length selected. The minimum and maximum total extension header length selected. The
maximum total header length is determined by the bits allocated for maximum total header length is determined by the bits allocated for
the total extension header length field. The risk with this approach the total extension header length field. The risk with this approach
is that it may be difficult to extend the total header size in the is that it may be difficult to extend the total header size in the
future. The minimum total header length is determined by a future. The minimum total header length is determined by a
requirement in the specifications that all implementations must meet. requirement in the specifications that all implementations must meet.
The risk with this approach is that all implementations will only The risk with this approach is that all implementations will only
implement the minimum total header length which would then become the implement the minimum total header length which would then become the
de facto maximum total header length. The recommended minimum total de facto maximum total header length. The recommended minimum total
header length is 64 bytes. header length is 64 bytes.
Single Extension size should always be 4 bytes aligned. Single Extension size should always be 4 byte aligned.
Internet-Draft NVO3 Encapsulation Considerations
The maximum length of a single option should be large enough to meet The maximum length of a single option should be large enough to meet
the different extension use case requirements e.g. in-band telemetry the different extension use case requirements, e.g., in-band
and future use. telemetry and future use.
6.5. Extension Ordering 6.5. Extension Ordering
In order to support hardware nodes at the tunnel endpoint or at the To support hardware nodes at the tunnel endpoint or at a transit
transit that can process one or few extensions TLVs in TCAM. A device that can process one or a few extensions TLVs in TCAM, a
control plane in such a deployment can signal a capability to ensure control plane in such a deployment can signal a capability to ensure
a specific TLV will always appear in a specific order for example the a specific TLV will always appear in a specific order, for example
first one in the packet. the first one in the packet.
The order of the TLVs should be HW friendly for both the sender and The order of the TLVs should be hardware friendly for both the sender
the receiver and possibly the transit node too. and the receiver and possibly the transit device also.
Transit nodes doesn't participate in control plane communication Transit devices doesn't participate in control plane communication
between the end points and are not required to process the options between the end points and are not required to process the options;
however, if they do, they need to process only a small subset of however, if they do, they need to process only a small subset of
options that will be consumed by tunnel endpoints. options that will be consumed by tunnel endpoints.
6.6. TLV vs Bit Fields 6.6. TLV versus Bit Fields
If there is a well-known initial set of options that are likely to be If there is a well-known initial set of options that are likely to be
implemented in software and in hardware, it can be efficient to use implemented in software and in hardware, it can be efficient to use
the bit-field approach as in GUE. However, as described in section the bit-field approach as in GUE. However, as described in section
6.3, if options are added over time and different subsets of options 6.3, if options are added over time and different subsets of options
are likely to be implemented in different pieces of hardware, then it are likely to be implemented in different pieces of hardware, then it
would be hard for the IETF to specify which options should get the would be hard for the IETF to specify which options should get the
early bit fields. TLVs are a lot more flexible, which avoids the early bit fields. TLVs are a lot more flexible, which avoids the
need to determine the relative importance different options. need to determine the relative importance different options.
However, general TLV of arbitrary order, size, and repetition of the However, general TLV of arbitrary order, size, and repetition of the
same order is difficult to implement in hardware. A middle ground is same order is difficult to implement in hardware. A middle ground is
to use TLV with restrictions on the size and alignment, observing to use TLVs with restrictions on their size and alignment, observing
that individual TLVs can have a fixed length, and support in the that individual TLVs can have a fixed length, and support in the
control plane such that an NVE will only receive options that to control plane such that an NVE will only receive options that it
needs and implements. The control plane approach can potentially be needs and implements. The control plane approach can potentially be
used to control the order of the TLVs sent to a particular NVE. Note used to control the order of the TLVs sent to a particular NVE. Note
that transit devices are not likely to participate in the control that transit devices are not likely to participate in the control
plane hence to the extent that they need to participate in option plane; hence, to the extent that they need to participate in option
processing they need more effort, But transit devices would have processing, they need more effort. Transit devices would have issues
issues with future GUE bits being defined for future options as well. with future GUE bits being defined for future options as well.
A benefit of TLVs from a HW perspective is that they are self A benefit of TLVs from a hardware perspective is that they are self
describing i.e., all the information is in the TLV. In a Bit fields describing, i.e., all the information is in the TLV. In a Bit fields
approach the hardware needs to look up the bit to determine the approach the hardware needs to look up the bit to determine the
length of the data associated with the bit through some separate length of the data associated with the bit through some separate
Internet-Draft NVO3 Encapsulation Considerations
table, which would add hardware complexity. table, which would add hardware complexity.
There are use cases where multiple modules of software are running on There are use cases where multiple modules of software are running on
NVE. This can be modules such as a diagnostic module by one vendor an NVE. This can be modules such as a diagnostic module by one
that does packet sampling and another module from a different vendor vendor that does packet sampling and another module from a different
that does a firewall. Using a TLV format, it is easier to have vendor that does a firewall. Using a TLV format, it is easier to
different software modules process different TLVs, which could be have different software modules process different TLVs, which could
standard extensions or vendor specific extensions defined by the be standard extensions or vendor specific extensions defined by the
different vendors, without conflicting with each other. This can different vendors, without conflicting with each other. This can
help with hardware modularity as well. There are some help with hardware modularity as well. There are some
implementations with options that allows different software like mac implementations with options that allows different software, like MAC
learning and security handle different options. learning and security, to handle different options.
6.7. Control Plane Considerations 6.7. Control Plane Considerations
Given that we want to allow large flexibility and extensibility for Given that we want to allow considerable flexibility and
e.g. software NVEs, yet be able to support key extensions in less extensibility for, e.g., software NVEs, yet be able to support
flexible e.g. hardware NVEs, it is useful to consider the control important extensions in less flexible contexts such as hardware NVEs,
plane. By control plane in this context we mean both protocols such it is useful to consider the control plane. By control plane in this
as EVPN and others, and also deployment specific configuration. section we mean both protocols, such as EVPN and others, and
deployment specific configuration.
If each NVE can express in the control plane that they only care If each NVE can express in the control plane that they only care
about particular extensions (could be a single extension, or a few), about particular extensions (could be a single extension, or a few),
and the source NVEs only include requested extensions in the NVO3 and the source NVEs only include requested extensions in the NVO3
packets, then the target NVE can both use a simpler parser (e.g., a packets, then the target NVE can both use a simpler parser (e.g., a
TCAM might be usable to look for a single NVO3 extension) and the TCAM might be usable to look for a single NVO3 extension) and the
depth of the inner payload in the NVO3 packet will be minimized. depth of the inner payload in the NVO3 packet will be minimized.
Furthermore, if the target NVE cares about a few extensions and can Furthermore, if the target NVE cares about a few extensions and can
express in the control plane the desired order of those extensions in express in the control plane the desired order of those extensions in
the NVO3 packets, then it can provide useful functionality with the NVO3 packets, then it can provide useful functionality with
minimal hardware requirements. minimal hardware requirements.
Note that transit devices that are not aware of the NVO3 extensions Note that transit devices that are not aware of the NVO3 extensions
somewhat benefit from such an approach, since the inner payload is somewhat benefit from such an approach, since the inner payload is
less deep in the packet if no extraneous extensions are included in less deep in the packet if no extraneous extensions are included in
the packet. However, in general a transit device is not likely to the packet. In general, a transit device is not likely to
participate in the NVO3 control plane. (However, configuration participate in the NVO3 control plane. (However, configuration
mechanisms can take into account limitations of the transit devices mechanisms can take into account limitations of the transit devices
used in particular deployments.) used in particular deployments.)
Note that in this approach different NVEs could desire different Note that in this approach different NVEs could desire different
(sets of) extensions, which means that the source NVE needs to be extensions or sets of extensions, which means that the source NVE
able to place different sets of extensions in different NVO3 packets, needs to be able to place different sets of extensions in different
and perhaps in different order. It also assumes that underlay NVO3 packets, and perhaps in different order. It also assumes that
multicast or replication servers are not used together with NVO3 underlay multicast or replication servers are not used together with
extensions. NVO3 extensions.
Internet-Draft NVO3 Encapsulation Considerations
There is a need to consider mandatory extensions versus optional There is a need to consider mandatory extensions versus optional
extensions. Mandatory extensions require the receiver to drop the extensions. Mandatory extensions require the receiver to drop the
packet if the extension is unknown. A control plane mechanism can packet if the extension is unknown. A control plane mechanism can
prevent the need for dropping unknown extensions, since they would prevent the need for dropping unknown extensions, since they would
not be included to targets that do not support them. not be included to targets that do not support them.
The control planes defined today need to add the ability to describe The control planes defined today need to add the ability to describe
the different encapsulations. Thus perhaps EVPN, and any other the different encapsulations. Thus, perhaps EVPN and any other
control plane protocol that the IETF defines, should have a way to control plane protocol that the IETF defines should have a way to
enumerate the supported NVO3 extensions and their order. enumerate the supported NVO3 extensions and their order.
The WG should consider developing a separate draft on guidance for The WG should consider developing a separate draft on guidance for
option processing and control plane participation. This should option processing and control plane participation. This should
provide examples/guidance on range of usage models and deployments provide examples/guidance on range of usage models and deployments
scenarios for specific options and ordering that are relevant for scenarios for specific options and ordering that are relevant for
that specific deployment. This includes end points and middle boxes that specific deployment. This includes end points and middle boxes
using the options. So, having the control plane negotiate the using the options. So, having the control plane negotiate the
constraints is most appropriate and flexible way to address these constraints is the most appropriate and flexible way to address these
requirements. requirements.
6.8. Split NVE 6.8. Split NVE
If the working group sees a need for having the hosts send and If the working group sees a need for having the hosts send and
receive options in a split NVE case, this is possible using any of receive options in a split NVE case, this is possible using any of
the existing extensible encapsulations (Geneve, GUE, GPE+NSH) by the existing extensible encapsulations (Geneve, GUE, GPE+NSH) by
defining a way to carry those over other transports. NSH can already defining a way to carry those over other transports. NSH can already
be used over different transports. be used over different transports.
If we need to do this with other encapsulations it can be done by If we need to do this with other encapsulations it can be done by
defining an Ether type for other encapsulations so that it can be defining an Ether type for other encapsulations so that it can be
carried over Ethernet and 802.1Q. carried over Ethernet and 802.1Q.
If we need to carry other encapsulations over MPLS, it would require If we need to carry other encapsulations over MPLS, it would require
an EVPN control plane to signal that other encapsulation header + an EVPN control plane to signal that other encapsulation header +
options will be present in front of the L2 packet. The VNI can be options will be present in front of the L2 packet. The VNI can be
ignored in the header, and the MPLS label will be the one used to ignored in the header, and the MPLS label will be the one used to
identify the EVPN L2 instance. identify the EVPN L2 instance.
6.9. Larger VNI Considerations 6.9. Larger VNI Considerations
We discussed whether we should make VNI 32-bits or larger. The We discussed whether we should make the VNI 32-bits or larger. The
benefit of 24-bit VNI would be to avoid unnecessary changes with benefit of a 24-bit VNI would be to avoid unnecessary changes with
existing proposals and implementations that are almost all, if not existing proposals and implementations that are almost all, if not
all, are using 24-bit VNI. If we need a larger VNI, an extension can all, using 24-bit VNI. If we need a larger VNI, an extension can be
be used to support that. used to support that.
7. Design team recommendations Internet-Draft NVO3 Encapsulation Considerations
We concluded that Geneve is most suitable as a starting point for 7. Design Team Recommendations
We concluded that Geneve is most suitable as a starting point for a
proposed standard for network virtualization, for the following proposed standard for network virtualization, for the following
reasons: reasons:
1. We studied whether VNI should be in base header or in extensions 1. We studied whether VNI should be in the base header or in
and whether it should be 24-bit or 32-bit. The design team agreed extensions and whether it should be 24-bit or 32-bit. The design
that VNI is critical information for network virtualization and MUST team agreed that VNI is critical information for network
be present in all packets. Design team also agreed that 24-bit VNI virtualization and MUST be present in all packets. The design team
matches the existing widely used encapsulation format i.e. VxLAN and also agreed that a 24-bit VNI matches the existing widely used
NVGRE and hence more suitable to use going forward. encapsulation formats, i.e., VxLAN and NVGRE, and hence is more
suitable to use going forward.
2. Geneve has the total options length that allow skipping over the 2. The Geneve header has the total options length which allows
options for NIC offload operations, and will allow transit devices to skipping over the options for NIC offload operations and will allow
view flow information in the inner payload. transit devices to view flow information in the inner payload.
3. We considered the option of using NSH with VxLAN-GPE but given 3. We considered the option of using NSH [RFC8300] with VxLAN-GPE
that NSH is targeted at service chaining and contains service but given that NSH is targeted at service chaining and contains
chaining information, it is less suitable for the network service chaining information, it is less suitable for the network
virtualization use case. The other downside for VxLAN-GPE was lack virtualization use case. The other downside for VxLAN-GPE was lack
of header length in VxLAN-GPE and hence makes skipping over the of header length in VxLAN-GPE which makes skipping over the headers
headers to process inner payload more difficult. Total Option Length to process inner payload more difficult. Total Option Length is
is present in Geneve. It is not possible to skip any options in the present in Geneve. It is not possible to skip any options in the
middle with VxLAN-GPE. In principle a split between a base header middle with VxLAN-GPE. In principle a split between a base header
and a header with options is interesting (whether that options header and a header with options is interesting (whether that options header
is NSH or some new header without ties to a service path). We is NSH or some new header without ties to a service path). We
explored whether it would make sense to either use NSH for this, or explored whether it would make sense to either use NSH for this, or
define a new NVO3 options header. However, we observed that this define a new NVO3 options header. However, we observed that this
makes it slightly harder to find the inner payload since the length makes it slightly harder to find the inner payload since the length
field is not in the NVO3 header itself. Thus one more field would field is not in the NVO3 header itself. Thus, one more field would
have to be extracted to compute the start of the inner payload. have to be extracted to compute the start of the inner payload.
Also, if the experience with IPv6 extension headers is a guidance, Also, if the experience with IPv6 extension headers is a guide, there
there would be a risk that key pieces of hardware might not implement would be a risk that key pieces of hardware might not implement the
the options header, resulting in future calls to deprecate its use. options header, resulting in future calls to deprecate its use.
Making the options part of the base NVO3 header has less of those Making the options part of the base NVO3 header has less of those
issues. Even though the implementation of any particular option can issues. Even though the implementation of any particular option can
not be predicted ahead of time, the option mechanism and ability to not be predicted ahead of time, the option mechanism and ability to
skip the options is likely to be broadly implemented. skip the options is likely to be broadly implemented.
4. We compared the TLV vs Bit-fields style extension and it was 4. We compared the TLV vs Bit-fields style extension and it was
deemed that parsing both TLV and bit-fields is expensive and while deemed that parsing both TLV and bit-fields is expensive and while
bit-fields may be simpler to parse, it is also more restrictive and bit-fields may be simpler to parse, it is also more restrictive and
requires guessing which extensions will be widely implemented so they requires guessing which extensions will be widely implemented so they
can get early bit assignments, given that half the bits are already can get early bit assignments, given that half the bits are already
assigned in GUE, a widely deployed extension may appear in a flag assigned in GUE, a widely deployed extension may appear in a flag
extension, and this will require extra processing, to dig the flag extension, and this will require extra processing, to dig the flag
from the flag extension and then look for the extension itself. As from the flag extension and then look for the extension itself. Also
well Bit-fields are not flexible enough to address the requirements Bit-fields are not flexible enough to address the requirements from
from OAM, Telemetry and security extensions, for variable length
option and different subtypes of the same option. While TLV are more Internet-Draft NVO3 Encapsulation Considerations
OAM, Telemetry, and security extensions, for variable length option
and different subtypes of the same option. While TLV are more
flexible, a control plane can restrict the number of option TLVs as flexible, a control plane can restrict the number of option TLVs as
well the order and size of the TLVs to make it simpler for a well the order and size of the TLVs to make it simpler for a
dataplane implementation to handle. dataplane implementation to handle.
5. We briefly discussed multi-vendor NVE case, and the need to allow 5. We briefly discussed the multi-vendor NVE case, and the need to
vendors to put their own extensions in the NVE header. This is allow vendors to put their own extensions in the NVE header. This is
possible with TLVs. possible with TLVs.
6. We also agreed that the C bit in Geneve is helpful to allow 6. We also agreed that the C bit in Geneve is helpful to allow a
receiver NVE to easily decide whether to process options or not. For receiver NVE to easily decide whether to process options or not, for
example a UUID based packet trace and how an optional extension such example a UUID based packet trace, and how an optional extension such
as that can be ignored by receiver NVE and thus make it easy for NVE as that can be ignored by a receiver NVE and thus make it easy for
to skip over the options. Thus the C-bit remains as defined in NVE to skip over the options. Thus, the C-bit remains as defined in
Geneve. Geneve.
7. There are already some extensions that are being discussed (see 7. There are already some extensions that are being discussed (see
section 6.2) of varying sizes, by using Geneve option it is possible section 6.2) of varying sizes. By using Geneve option it is possible
to get in band parameters like: switch id, ingress port, egress port, to get in band parameters like switch id, ingress port, egress port,
internal delay, and queue in telemetry defined extension TLV from internal delay, and queue in telemetry defined extension TLV from
switches. It is also possible to add Security extension TLVs like switches. It is also possible to add Security extension TLVs like
HMAC and DTLS/IPSEC to authenticate the Geneve packet header and HMAC and DTLS/IPSEC to authenticate the Geneve packet header and
secure the Geneve packet payload by software or hardware tunnel secure the Geneve packet payload by software or hardware tunnel
endpoints. As well, a Group Based Policy extension TLV can be endpoints. A Group Based Policy extension TLV can be carried as
carried. well.
8. There are implemented Geneve options today in production. There 8. There are implemented Geneve options today in production. There
are as well new HW supporting Geneve TLV parsing. In addition In- are as well new hardware supporting Geneve TLV parsing. In addition,
band Telemetry (INT) specification being developed by P4.org an In-band Telemetry (INT) specification is being developed by P4.org
illustrates the option of INT meta data carried over Geneve. OVN/OVS that illustrates the option of INT meta data carried over Geneve.
have also defined some option TLV(s) for Geneve. OVN/OVS have also defined some option TLV(s) for Geneve.
9. The DT has addressed the usage models while considering the 9. The DT has addressed the usage models while considering the
requirements and implementations in general that includes software requirements and implementations in general that includes software
and hardware. and hardware.
There seems to be interest to standardize some well known secure There seems to be interest to standardize some well-known secure
option TLVs to secure the header and payload to guarantee option TLVs to secure the header and payload to guarantee
encapsulation header integrity and tenant data privacy. The design encapsulation header integrity and tenant data privacy. The design
team recommends that the working group consider standardizing such team recommends that the working group consider standardizing such
option(s). option(s).
We recommend the following enhancements to Geneve to make it more We recommend the following enhancements to Geneve to make it more
suitable to hardware and yet provide the flexibility for software: suitable to hardware and yet provide the flexibility for software:
We would propose a text such as, while TLV are more flexible, a We would propose a text such as, while TLV are more flexible, a
control plane can restrict the number of option TLVs as well the control plane can restrict the number of option TLVs as well the
order and size of the TLVs to make it simpler for a data plane order and size of the TLVs to make it simpler for a data plane
implementation in software or hardware to handle. For example, there implementation in software or hardware to handle. For example, there
may be some critical information such as secure hash that must be
processed in certain order at lowest latency. Internet-Draft NVO3 Encapsulation Considerations
may be some critical information such as a secure hash that must be
processed in a certain order at lowest latency.
A control plane can negotiate a subset of option TLVs and certain TLV A control plane can negotiate a subset of option TLVs and certain TLV
ordering, as well can limit the total number of option TLVs present ordering, as well as limiting the total number of option TLVs present
in the packet, for example, to allow hardware capable of processing in the packet, for example, to allow for hardware capable of
fewer options. Hence, the control planes need to have the ability to processing fewer options. Hence, the control plane needs to have the
describe the supported TLVs subset and their order. ability to describe the supported TLVs subset and their order.
The Geneve draft could specify that the subset and order of option The Geneve draft could specify that the subset and order of option
TLVs should be configurable for each remote NVE in the absence of a TLVs should be configurable for each remote NVE in the absence of a
protocol control plane. protocol control plane.
We recommend Geneve to follow fragmentation recommendations in We recommend that Geneve follow fragmentation recommendations in
overlay services like PWE3, and L2/L3 VPN recommendation to guarantee overlay services like PWE3 and the L2/L3 VPN recommendations to
larger MTU for the tunnel overhead [RFC3985],Section 5.3. guarantee larger MTU for the tunnel overhead ([RFC3985] Section 5.3).
We request Geneve to provide a recommendation for critical bit We request that Geneve provide a recommendation for critical bit
processing - text could look like how critical bits can be used with processing - text could specify how critical bits can be used with
control plane specifying the critical options. control plane specifying the critical options.
Given that there is a telemetry option use case for a length of 256 Given that there is a telemetry option use case for a length of 256
bytes, we recommend Geneve to increase the Single TLV option length bytes, we recommend that Geneve increase the Single TLV option length
to 256. to 256.
We request Geneve to address Requirements for OAM considerations for We request that Geneve address Requirements for OAM considerations
alternate marking and for performance measurements that need 2 bits for alternate marking and for performance measurements that need 2
in the header. And clarify the need of the current OAM bit in the bits in the header and clarify the need for the current OAM bit in
Geneve Header. the Geneve Header.
We recommend the WG to work on security options for Geneve. We recommend that the WG work on security options for Geneve.
8. Acknowledgements Internet-Draft NVO3 Encapsulation Considerations
8. Acknowledgements
The authors would like to thank Tom Herbert for providing the The authors would like to thank Tom Herbert for providing the
motivation for the Security/Integrity extension, and for his valuable motivation for the Security/Integrity extension, and for his valuable
comments, and would like to thank T. Sridhar for his valuable comments, and would like to thank T. Sridhar for his valuable
comments and feedback. comments and feedback.
9. Security Considerations 9. Security Considerations
This document does not introduce any additional security constraints. This document does not introduce any additional security constraints.
10. IANA Considerations 10. IANA Considerations
This document has no actions for IANA. This document has no actions for IANA.
11. Appendix A Internet-Draft NVO3 Encapsulation Considerations
11.1. Overview 11. References
11.1 Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, DOI
10.17487/RFC2119, March 1997, <https://www.rfc-
editor.org/info/rfc2119>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119
Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May
2017, <https://www.rfc-editor.org/info/rfc8174>.
11.2 Informative References
[I-D.herbert-gue-extensions] Herbert, T., Yong, L., and F. Templin,
"Extensions for Generic UDP Encapsulation",
draft-herbert-gue-extensions-01 (work in progress), October
2016.
[I-D.ietf-intarea-gue] Herbert, T., Yong, L., and O. Zia, "Generic
UDP Encapsulation", draft-ietf-intarea-gue (work in
progress), October 2019.
[I-D.ietf-nvo3-vxlan-gpe] Maino, F., Kreeger, L., and U. Elzur,
"Generic Protocol Extension for VXLAN",
draft-ietf-nvo3-vxlan-gpe (work in progress), March 2021.
[I-D.smith-vxlan-group-policy] Smith, M. and L. Kreeger, "VXLAN Group
Policy Option", draft-smith-vxlan-group-policy-05 (work in
progress), October 2018.
[RFC2418] Bradner, S., "IETF Working Group Guidelines and
Procedures", BCP 25, RFC 2418, DOI 10.17487/RFC2418,
September 1998, <https://www.rfc-editor.org/info/rfc2418>.
[RFC3985] Bryant, S., Ed. and P. Pate, Ed., "Pseudo Wire Emulation
Edge-to-Edge (PWE3) Architecture", RFC 3985, DOI
10.17487/RFC3985, March 2005, <https://www.rfc-
editor.org/info/rfc3985>.
[RFC8300] Quinn, P., Ed., Elzur, U., Ed., and C. Pignataro, Ed.,
"Network Service Header (NSH)", RFC 8300, DOI
10.17487/RFC8300, January 2018, <https://www.rfc-
editor.org/info/rfc8300>.
Internet-Draft NVO3 Encapsulation Considerations
[RFC8926] Gross, J., Ed., Ganga, I., Ed., and T. Sridhar, Ed.,
"Geneve: Generic Network Virtualization Encapsulation", RFC
8926, DOI 10.17487/RFC8926, November 2020,
<https://www.rfc-editor.org/info/rfc8926>.
Internet-Draft NVO3 Encapsulation Considerations
Appendix A: Encapsulations Comparison
A.1. Overview
This section presents a comparison of the three NVO3 encapsulation This section presents a comparison of the three NVO3 encapsulation
proposals, Geneve, GUE, and VXLAN-GPE. The three encapsulations use proposals, Geneve, GUE, and VXLAN-GPE. The three encapsulations use
an outer UDP/IP transport. Geneve and VXLAN-GPE use an 8-octet an outer UDP/IP transport. Geneve and VXLAN-GPE use an 8-octet
header, while GUE uses a 4-octet header. In addition to the base header, while GUE uses a 4-octet header. In addition to the base
header, optional extensions may be included in the encapsulation, as header, optional extensions may be included in the encapsulation, as
discussed in Section 3.2 below. discussed in Section A.2 below.
11.2. Extensibility A.2. Extensibility
11.2.1. Native Extensibility Support A.2.1. Native Extensibility Support
The Geneve and GUE encapsulations both enable optional headers to be The Geneve and GUE encapsulations both enable optional headers to be
incorporated at the end of the base encapsulation header. incorporated at the end of the base encapsulation header.
VXLAN-GPE does not provide native support for header extensions. VXLAN-GPE does not provide native support for header extensions.
However, as discussed in [I-D.ietf-nvo3-vxlan-gpe], extensibility can However, as discussed in [I-D.ietf-nvo3-vxlan-gpe], extensibility can
be attained to some extent if the Network Service Header (NSH) be attained to some extent if the Network Service Header (NSH)
[RFC8300] is used immediately following the VXLAN-GPE header. NSH [RFC8300] is used immediately following the VXLAN-GPE header. NSH
supports either a fixed-size extension (MD Type 1), or a variable- supports either a fixed-size extension (MD Type 1), or a variable-
size TLV-based extension (MD Type 2). It should be noted that NSH- size TLV-based extension (MD Type 2). It should be noted that NSH-
over-VXLAN-GPE implies an additional overhead of the 8- octets NSH over-VXLAN-GPE implies an additional overhead of the 8-octets NSH
header, in addition to the VXLAN-GPE header. header, in addition to the VXLAN-GPE header.
11.2.2. Extension Parsing A.2.2. Extension Parsing
The Geneve Variable Length Options are defined as Type/Length/ The Geneve Variable Length Options are defined as Type/Length/Value
Value(TLV) extensions. Similarly, VXLAN-GPE, when using NSH, can (TLV) extensions. Similarly, VXLAN-GPE, when using NSH, can include
include NSH TLV-based extensions. In contrast, GUE defines a small NSH TLV-based extensions. In contrast, GUE defines a small set of
set of possible extension fields (proposed in possible extension fields (proposed in [I-D.herbert-gue-extensions]),
[I-D.herbert-gue-extensions], and a set of flags in the GUE header and a set of flags in the GUE header that indicate for each extension
that indicate for each extension type whether it is present or not. type whether it is present or not.
TLV-based extensions, as defined in Geneve, provide the flexibility TLV-based extensions, as defined in Geneve, provide the flexibility
for a large number of possible extension types. Similar behavior can for a large number of possible extension types. Similar behavior can
be supported in NSH-over-VXLAN-GPE when using MD Type 2. The flag- be supported in NSH-over-VXLAN-GPE when using MD Type 2. The flag-
based approach taken in GUE strives to simplify implementations by based approach taken in GUE strives to simplify implementations by
defining a small number of possible extensions, used in a fixed defining a small number of possible extensions used in a fixed order.
order.
Internet-Draft NVO3 Encapsulation Considerations
The Geneve and GUE headers both include a length field, defining the The Geneve and GUE headers both include a length field, defining the
total length of the encapsulation, including the optional extensions. total length of the encapsulation, including the optional extensions.
The length field simplifies the parsing of transit devices that skip The length field simplifies the parsing of transit devices that skip
the encapsulation header without parsing its extensions. the encapsulation header without parsing its extensions.
11.2.3. Critical Extensions A.2.3. Critical Extensions
The Geneve encapsulation header includes the 'C' field, which The Geneve encapsulation header includes the 'C' field, which
indicates whether the current Geneve header includes critical indicates whether the current Geneve header includes critical
options, which must be parsed by the tunnel endpoint. If the options, that is to say, options which must be parsed by the tunnel
endpoint is not able to process the critical option, the packet is endpoint. If the endpoint is not able to process a critical option,
discarded. the packet is discarded.
11.2.4. Maximal Header Length A.2.4. Maximal Header Length
The maximal header length in Geneve, including options, is 260 The maximal header length in Geneve, including options, is 260
octets. GUE defines the maximal header to be 128 octets. VXLAN-GPE octets. GUE defines the maximal header to be 128 octets. VXLAN-GPE
uses a fixed-length header of 8 octets, unless NSH-over-VXLAN-GPE is uses a fixed-length header of 8 octets, unless NSH-over-VXLAN-GPE is
used, yielding an encapsulation header of up to 264 octets. used, yielding an encapsulation header of up to 264 octets.
11.3. Encapsulation Header A.3. Encapsulation Header
11.3.1. Virtual Network Identifier (VNI) A.3.1. Virtual Network Identifier (VNI)
The Geneve and VXLAN-GPE headers both include a 24-bit VNI field. The Geneve and VXLAN-GPE headers both include a 24-bit VNI field.
GUE, on the other hand, enables the use of a 32-bit field called GUE, on the other hand, enables the use of a 32-bit field called
VNID; this field is not included in the GUE header, but was defined VNID; this field is not included in the GUE header, but was defined
as an optional extension in [I-D.herbert-gue-extensions]. as an optional extension in [I-D.herbert-gue-extensions].
The VXLAN-GPE header includes the 'I' bit, indicating that the VNI The VXLAN-GPE header includes the 'I' bit, indicating that the VNI
field is valid in the current header. A similar indicator is defined field is valid in the current header. A similar indicator is defined
as a flag in the GUE header herbert-gue-extensions. as a flag in the GUE header [I-D.herbert-gue-extensions].
11.3.2. Next Protocol A.3.2. Next Protocol
The three encapsulation headers include a field that specifies the The three encapsulation headers include a field that specifies the
type of the next protocol header, which resides after the NVO3 type of the next protocol header, which resides after the NVO3
encapsulation header. The Geneve header includes a 16-bit field that encapsulation header. The Geneve header includes a 16-bit field that
uses the IEEE Ethertype convention. GUE uses an 8-bit field, which uses the IEEE Ethertype convention. GUE uses an 8-bit field, which
Internet-Draft NVO3 Encapsulation Considerations
uses the IANA Internet protocol numbering. The VXLAN-GPE header uses the IANA Internet protocol numbering. The VXLAN-GPE header
incorporates an 8-bit Next Protocol field, using a VXLAN-GPE-specific incorporates an 8-bit Next Protocol field, using a VXLAN-GPE-specific
registry, defined in [I-D.ietf-nvo3-vxlan-gpe]. registry, defined in [I-D.ietf-nvo3-vxlan-gpe].
The VXLAN-GPE header also includes the 'P' bit, which explicitly The VXLAN-GPE header also includes the 'P' bit, which explicitly
indicates whether the Next Protocol field is present in the current indicates whether the Next Protocol field is present in the current
header. header.
11.3.3. Other Header Fields A.3.3. Other Header Fields
The OAM bit, which is defined in Geneve and in VXLAN-GPE, indicates The OAM bit, which is defined in Geneve and in VXLAN-GPE, indicates
whether the current packet is an OAM packet. The GUE header includes whether the current packet is an OAM packet. The GUE header includes
a similar field, but uses different terminology; the GUE 'C-bit' a similar field, but uses different terminology; the GUE 'C-bit'
specifies whether the current packet is a control packet. Note that specifies whether the current packet is a control packet. Note that
the GUE control bit can potentially be used in a large set of the GUE control bit can potentially be used in a large set of
protocols that are not OAM protocols. However, the control packet protocols that are not OAM protocols. However, the control packet
examples discussed in [I-D.ietf-nvo3-gue] are OAM-related. examples discussed in [I-D.ietf-intarea-gue] are OAM-related.
Each of the three NVO3 encapsulation headers includes a 2-bit Version Each of the three NVO3 encapsulation headers includes a 2-bit Version
field, which is currently defined to be zero. field, which is currently defined to be zero.
The Geneve and VXLAN-GPE headers include reserved fields; 14 bits in The Geneve and VXLAN-GPE headers include reserved fields; 14 bits in
the Geneve header, and 27 bits in the VXLAN-GPE header are reserved. the Geneve header, and 27 bits in the VXLAN-GPE header are reserved.
11.4. Comparison Summary A.4. Comparison Summary
Internet-Draft NVO3 Encapsulation Considerations
The following table summarizes the comparison between the three NVO3 The following table summarizes the comparison between the three NVO3
encapsulations. encapsulations:
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| | Geneve | GUE | VXLAN-GPE | | | Geneve | GUE | VXLAN-GPE |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| Outer transport| UDP/IP | UDP/IP | UDP/IP | | Outer transport| UDP/IP | UDP/IP | UDP/IP |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| Base header | 8 octets | 4 octets | 8 octets | | Base header | 8 octets | 4 octets | 8 octets |
| length | | | (16 octets | | length | | | (16 octets |
| | | | using NSH) | | | | | using NSH) |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| Extensibility |Variable length |Extension fields| No native ext- | | Extensibility |Variable length |Extension fields| No native ext- |
skipping to change at page 17, line 31 skipping to change at page 22, line 55
| indicator | | | | | indicator | | | |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| OAM / control | OAM bit | Control bit | OAM bit | | OAM / control | OAM bit | Control bit | OAM bit |
| field | | | | | field | | | |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| Version field | 2 bits | 2 bits | 2 bits | | Version field | 2 bits | 2 bits | 2 bits |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
| Reserved bits | 14 bits | - | 27 bits | | Reserved bits | 14 bits | - | 27 bits |
+----------------+----------------+----------------+----------------+ +----------------+----------------+----------------+----------------+
Figure 1: NVO3 Encapsulation Comparison Figure 1: NVO3 Encapsulations Comparison
12. Contributors
the following co-authors have contributed to this document.
Ilango Ganga Intel Email: ilango.s.ganga@intel.com
Pankaj Garg Microsoft Email: pankajg@microsoft.com
Rajeev Manur Broadcom Email: rajeev.manur@broadcom.com
Tal Mizrahi Marvell Email: talmi@marvell.com
David Mozes Email: mosesster@gmail.com
Erik Nordmark Email: nordmark@sonic.net Internet-Draft NVO3 Encapsulation Considerations
Michael Smith Cisco Email: michsmit@cisco.com Contributors
Sam Aldrin Google Email: aldrin.ietf@gmail.com
Ignas Bagdonas Equinix Email: ibagdona.ietf@gmail.com The following co-authors have contributed to this document:
13. References Ilango Ganga Intel Email: ilango.s.ganga@intel.com
13.1. Normative References Pankaj Garg Microsoft Email: pankajg@microsoft.com
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Rajeev Manur Broadcom Email: rajeev.manur@broadcom.com
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
13.2. Informative References Tal Mizrahi Marvell Email: talmi@marvell.com
[I-D.herbert-gue-extensions] David Mozes Email: mosesster@gmail.com
Herbert, T., Yong, L., and F. Templin, "Extensions for
Generic UDP Encapsulation", draft-herbert-gue-
extensions-01 (work in progress), October 2016.
[I-D.ietf-nvo3-geneve] Erik Nordmark Email: nordmark@sonic.net
Gross, J., Ganga, I., and T. Sridhar, "Geneve: Generic
Network Virtualization Encapsulation", draft-ietf-
nvo3-geneve-14 (work in progress), September 2019.
[I-D.ietf-nvo3-gue] Michael Smith Cisco Email: michsmit@cisco.com
Herbert, T., Yong, L., and O. Zia, "Generic UDP
Encapsulation", draft-ietf-nvo3-gue-05 (work in progress),
October 2016.
[I-D.ietf-nvo3-vxlan-gpe] Sam Aldrin Google Email: aldrin.ietf@gmail.com
Maino, F., Kreeger, L., and U. Elzur, "Generic Protocol
Extension for VXLAN", draft-ietf-nvo3-vxlan-gpe-09 (work
in progress), December 2019.
[I-D.smith-vxlan-group-policy] Ignas Bagdonas Equinix Email: ibagdona.ietf@gmail.com
Smith, M. and L. Kreeger, "VXLAN Group Policy Option",
draft-smith-vxlan-group-policy-05 (work in progress),
October 2018.
[RFC2418] Bradner, S., "IETF Working Group Guidelines and Internet-Draft NVO3 Encapsulation Considerations
Procedures", BCP 25, RFC 2418, DOI 10.17487/RFC2418,
September 1998, <https://www.rfc-editor.org/info/rfc2418>.
[RFC3985] Bryant, S., Ed. and P. Pate, Ed., "Pseudo Wire Emulation Authors' Addresses
Edge-to-Edge (PWE3) Architecture", RFC 3985,
DOI 10.17487/RFC3985, March 2005,
<https://www.rfc-editor.org/info/rfc3985>.
[RFC8300] Quinn, P., Ed., Elzur, U., Ed., and C. Pignataro, Ed., Sami Boutros (editor)
"Network Service Header (NSH)", RFC 8300, Ciena
DOI 10.17487/RFC8300, January 2018, USA
<https://www.rfc-editor.org/info/rfc8300>.
Author's Address Email: sboutros@ciena.com
Sami Boutros (editor) Donald E. Eastlake, 3rd (editor)
Ciena Futurewei Technologies
USA 2386 Panoramic Circle
Apopka, FL 32703
USA
Email: sboutros@ciena.com Tel: +1-508-333-2270
Email: d3e3e3@gmail.com
 End of changes. 161 change blocks. 
332 lines changed or deleted 413 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/