NVO3 WG T. Ao
Internet-Draft ZTE Corporation
Intended status: Standards Track G. Mirsky
Expires: August 31, 2018 ZTE Corp.
Y. Fan
China Telecom
February 27, 2018
Multi-encapsulation interconnection for Overlay Virtual Network
draft-ao-nvo3-multi-encap-interconnect-00
Abstract
For an virtualized overlay network, there are many encapsulations
that may be used. Different customer have their own preference. So
if some of these different encapsulation can be interconnected
together, the virtualized overlay network would be more compatible
and have loose strict on access. This document is going to provide
an architecture of different overlay encapsulation interconnection
and an tranformer gateway for these end station connected to the
virtual network.
Status of This Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79.
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 https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
This Internet-Draft will expire on August 31, 2018.
Copyright Notice
Copyright (c) 2018 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(https://trustee.ietf.org/license-info) in effect on the date of
Ao, et al. Expires August 31, 2018 [Page 1]
Internet-Draft multiple encapsulation interconnect February 2018
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Conventions used in this document . . . . . . . . . . . . . . 3
2.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 3
2.2. Requirements Language . . . . . . . . . . . . . . . . . . 3
3. Multi-encapsulation NVO architecture . . . . . . . . . . . . 3
3.1. Transformer NVE . . . . . . . . . . . . . . . . . . . . . 5
4. Control Plane . . . . . . . . . . . . . . . . . . . . . . . . 6
4.1. NVE to NVA . . . . . . . . . . . . . . . . . . . . . . . 6
4.2. NVA to NVE . . . . . . . . . . . . . . . . . . . . . . . 7
4.3. NVA to NVA . . . . . . . . . . . . . . . . . . . . . . . 7
5. Security Considerations . . . . . . . . . . . . . . . . . . . 7
6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7
7. References . . . . . . . . . . . . . . . . . . . . . . . . . 7
7.1. Normative References . . . . . . . . . . . . . . . . . . 7
7.2. Informational References . . . . . . . . . . . . . . . . 8
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9
1. Introduction
Network virtualization using Overlays over Layer 3 (NVO3) is a
technology that is used to address issues that arise in building
large, multi-tenant data centers that make extensive use of server
virtualization.
With the progress in NVO3 WG, some of the data plane encapsulations
have been put forward, some are outstanding dataplane for overlay
network, such as VxLAN-GPE [I-D.ietf-nvo3-vxlan-gpe], GENEVE
[I-D.ietf-nvo3-geneve] and GUE [I-D.ietf-nvo3-encap], etc. The
consideration about these overlay encapsulations has been analysised
in [I-D.ietf-nvo3-encap]. The fact is that each of them have its
custmers, and furthermore, some of them have been provisioned in the
network. So that a problem arises: for a virtual network, all the
hosts that connect to the same VN and want to communicate with each
other are required to have the same data plane encapsulation. This
problem limite the network scalability and capacity. Especially,
when the NVE is located on the vSwitch, the encapsulation method on
the NVE is not predictable. Allowing as many kinds of accession as
possible is more attractive for a virtualized overlay network.
Ao, et al. Expires August 31, 2018 [Page 2]
Internet-Draft multiple encapsulation interconnect February 2018
To improve the scalability and capacity of the virtualized overlay
network, we propose a multi-encapsulation access allowed interconnect
NVO3 architecture, and a gateway in the network to perform the
transformation for different encapsulation in this document, by which
these hosts with different encapsulations can be interconnected.
Here we call the gateway as Transformer Gateway in the following
descrption.
2. Conventions used in this document
2.1. Terminology
NVO3: Network Virtualization using Overlay over Layer 3
NVA: Network Virtualization Authority
TS: Tenant System
VxLAN-GPE: Virtual extension LAN with Generic Protocol Extension
GENEVE: Generic Network Virtualization Encapsulation
GUE: Generic UDP Encapsulation
Multi-encapsulation NVO: an virtualized overlay network that allow
multiple different encapsulations interconnection.
t-GW: Tranformation Gateway. A gateway that do the transformer for
different encapsulation to make them can communicate with each other.
tNVE: A NVE that complete the functionn of a tranformer gateway.
2.2. Requirements Language
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"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.
3. Multi-encapsulation NVO architecture
In the multi-encapsulation interconnection allowed NVO, different NVE
may support different encapsulation data plane. As we have know that
any two of the TS in the same VN should communicate with each other,
but it is required that both of overaly encapsulation they are using
to access to the virtual network have to be same. In order to
relieve the limitation and to support these encapsulation would
Ao, et al. Expires August 31, 2018 [Page 3]
Internet-Draft multiple encapsulation interconnect February 2018
interconnect together, a multi-encapsulation architecture is
introdueced. Figure 1 dipcts a reference architecture in VN.
+--------+
| Tenant +--+
| System4| |
+--------+ | ..................
| +---+ +---------------+
+--|NVE|---+ +---|Transformer NVE|
+---+ | | | (tNVE) |
/. | | +---------------+
/ . +-----+ .
/ . +--| NVA |--+ .
/ . | +-----+ \ .
| . | \ .
| . | Overlay +--+--++--------+
+--------+ | . | Network | NVE || Tenant |
| Tenant +--+ . | | || System5|
| System3| . \ +---+ +--+--++--------+
+--------+ .....|NVE|.........
+---+
|
|
=====================
| |
+--------+ +--------+
| Tenant | | Tenant |
| System1| | System2|
+--------+ +--------+
Figure 1 Multi-encapsulation VN architecture
Figure 1: Multi-encapsulation interconnection architecture
In this figure, a Transformer Gateway component is introduced.
Generally, the gateway is located on a NVE, so we call it as tNVE.
For the TSs in the same virtual network, if the NVE they connecting
have different encapsulation, want to communicate with each other,
Transformer NVE(tNVE) will take over as a gateway to provide a
"bridge" for the communication. That is, when different NVEs want to
set up tunnel, if they can't connect each other directly because of
different encapsulation, they can set up a tunnel with tNVE
seperately, so that the tNVE connects the two tunnels as a transfer.
There could be more than one tNVE in a network.
Ao, et al. Expires August 31, 2018 [Page 4]
Internet-Draft multiple encapsulation interconnect February 2018
3.1. Transformer NVE
Transformer NVE(tNVE) is a certain kind of NVE that maybe appointed
by NVA or by manager. As a very important component in the multi-
encapsulation NVO, one requirement for NVE being a Transfomer NVE is
that the NVE should support at least two kinds of encapsulations.
With reference to the [RFC8014], the Transformer NVE has a reference
model as showed in Figure 2.
| |
| Data-Center Network (IP) |
+-----------------------------------------------------------+
| | | |
| Tunnel Overlay | | Tunnel Overlay |
+---------+----+ +--+----------+----+ +-----+--------+
| +----------+ | | +--------------+ | | +----------+ |
| | Overlay | | | | Overlay | | | | Overlay | |
| | Module | | | | Module | | | | Module | |
| +----+-----+ | | +---+----------+ | | +----------+ |
| | | | | | | | | |
| NVE1 | | | tNVE| | | | NVE3 | |
| +---+----+ | | +---+-+ +--+--+ | | +--------+ |
| | VNI1 | | | |VNI1 | |VNI1 | | | | VNI1 | |
| +-+------+ | | +-+---+ +-----+ | | +-+------+ |
| | VAP1 | | |VAP1 |VAP2| | | VAP1 |
+----+---------+ | +---------+ | +----+---------+
| +------------------+ |
|\ |
-----+-\------------------------------------------------------+-------
TSI1 |TSI2\ Tenant TSI1 |
+---+ +---+ +---+
|TS1| |TS2| |TS3|
+---+ +---+ +---+
Figure 2 tNVE Reference Model
Figure 2: tNVE reference model
tNVE is a key component for the connection between NVE1 and NVE2. It
can be a dedicated device and be a NVE that also provide the overlay
network for the TSs. When the NVE takes the role of transform
different encapsulation for different TSs, it will not forward the
traffic to TS, but to another VAP that support the encapsulation the
destination NVE owned.
Take the Figure 2 as an example to illustrate how does tNVE work.
NVE1 only support VxLAN-GPE, and NVE2 only support GENEVE. For the
two communiting TSs: TS1 wants to send packets to TS3, and TS3 also
Ao, et al. Expires August 31, 2018 [Page 5]
Internet-Draft multiple encapsulation interconnect February 2018
wants to reply to TS1. They are in the same VNID1, but the NVE they
are connecting using different encapsulation. So if the two TS
communicate with each other, packets have to transfer at tNVE. For
NVE1, it has no sense that TS3 is connecting to NVE3, instead
assuming that TS3 is connecting to tNVE. In the same way, for NVE3,
it has no sense that TS1 is connecting to NVE1, instead of assuming
that TS1 is connecting to tNVE. So because of the existence of the
tNVE, no matter TS1/TS3 or NVE1/NVE3, they never perceive that they
are in different data plane. NVE1 getting the packets from TS1
encapsulates them in Vxlan-GPE and then send the packets to tNVE.
The tNVE gets the packets from the Vxlan-GPE tunnel and then de-
encapsulate the vxLAN-GPE to VAP1. Next the tNVE forward packets to
the Overlay Module from VAP2 to have another encapsulation GENEVE on
the packets. At last tNVE forward the packet in the GENEVE tunnel to
NVE3.
From the above, tNVE is like a tranformer between TS1 and TS2. And
owning to tNVE, even though NVE1 and NVE2 which TS1 and TS2
connecting seperately have different encapsulation, as long as they
are in the same virtual network, they would communicate each other
and no need to have knowledge that they are in different data plane.
4. Control Plane
As stated in [RFC8014], an NVA is the entity that provides address
mapping and other information to NVEs. In addition, information
flows between NVEs and NVAs in both directions. The NVA maintains
information about all VNs in the NV Domain so that NVEs do not need
to do so themselves. NVEs obtain information from the NVA about
where a given remote TS destination resides. NVAs, in turn, obtain
information from NVEs about the individual TSs attached to those
NVEs.
Therefore, in order to make the tNVE works properly and to make sure
that all the other network entities except tNVE don't detect the
difference, NVA should detect the role of tNVE and take the
coordination role among tNVE and other NVEs, maintain the information
about the tNVEs and other NVEs, compute the tunnel connection, and
notify tNVEs and NVEs about remote TS information.
4.1. NVE to NVA
NVE and tNVE should not only notify the NVA the address mapping
between NVE and TSs, but also notify the NVA which encapsulation
tunnel does this mapping use, so that NVA be able to decide which
connection need tNVE participation.
Information from tNVE to NVA:
Ao, et al. Expires August 31, 2018 [Page 6]
Internet-Draft multiple encapsulation interconnect February 2018
1. Encapsulations the tNVE support.
Information from NVE to NVA:
1. The address mapping between NVE and its attached TSs.
2. The encapsulation tunnel the address mapping will use.
3. The mandate metadata the address mapping must use.
4.2. NVA to NVE
NVA notify the NVE and tNVE the mapping information after computing
and coordination.
Information from NVA to NVE:
1. The address mapping information between remote NVE and TS. The
NVE here may not be the NVE that the TS is connecting. It may be a
tNVE.
Information from NVA to tNVE:
1. The address mapping information between remote NVE and its
connecting TS.
4.3. NVA to NVA
For NVA federate scenario. To be added in the future updates.
5. Security Considerations
Will be added in the future updates.
6. IANA Considerations
TBD.
7. References
7.1. Normative References
[I-D.ietf-intarea-gue]
Herbert, T., Yong, L., and O. Zia, "Generic UDP
Encapsulation", draft-ietf-intarea-gue-05 (work in
progress), December 2017.
Ao, et al. Expires August 31, 2018 [Page 7]
Internet-Draft multiple encapsulation interconnect February 2018
[I-D.ietf-nvo3-geneve]
Gross, J., Ganga, I., and T. Sridhar, "Geneve: Generic
Network Virtualization Encapsulation", draft-ietf-
nvo3-geneve-05 (work in progress), September 2017.
[I-D.ietf-nvo3-vxlan-gpe]
Maino, F., Kreeger, L., and U. Elzur, "Generic Protocol
Extension for VXLAN", draft-ietf-nvo3-vxlan-gpe-05 (work
in progress), October 2017.
[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>.
[RFC7365] Lasserre, M., Balus, F., Morin, T., Bitar, N., and Y.
Rekhter, "Framework for Data Center (DC) Network
Virtualization", RFC 7365, DOI 10.17487/RFC7365, October
2014, <https://www.rfc-editor.org/info/rfc7365>.
[RFC8014] Black, D., Hudson, J., Kreeger, L., Lasserre, M., and T.
Narten, "An Architecture for Data-Center Network
Virtualization over Layer 3 (NVO3)", RFC 8014,
DOI 10.17487/RFC8014, December 2016,
<https://www.rfc-editor.org/info/rfc8014>.
[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>.
7.2. Informational References
[I-D.ietf-nvo3-encap]
Boutros, S., Ganga, I., Garg, P., Manur, R., Mizrahi, T.,
Mozes, D., Nordmark, E., Smith, M., Aldrin, S., and I.
Bagdonas, "NVO3 Encapsulation Considerations", draft-ietf-
nvo3-encap-01 (work in progress), October 2017.
[RFC7364] Narten, T., Ed., Gray, E., Ed., Black, D., Fang, L.,
Kreeger, L., and M. Napierala, "Problem Statement:
Overlays for Network Virtualization", RFC 7364,
DOI 10.17487/RFC7364, October 2014,
<https://www.rfc-editor.org/info/rfc7364>.
Ao, et al. Expires August 31, 2018 [Page 8]
Internet-Draft multiple encapsulation interconnect February 2018
Authors' Addresses
Ting Ao
ZTE Corporation
No.889, BiBo Road
Shanghai 201203
China
Phone: +86 21 68897642
Email: ao.ting@zte.com.cn
Greg Mirsky
ZTE Corp.
1900 McCarthy Blvd. #205
Milpitas, CA 95035
USA
Email: gregimirsky@gmail.com
Yongbin
China Telecom
No.109, Zhongshan Avenue
Guangzhou 510630
China
Email: fanyb@gsta.com
Ao, et al. Expires August 31, 2018 [Page 9]