Carrying Big Labels in BGP-4
draft-renwei-mpls-bgp-big-label-00.txt

Abstract

When BGP is used to distribute a particular route, it can also be used to distribute an MPLS label which is mapped to that route. In some cases, for example, when L3VPN is used to access and connect to virtual networks in data centers, there may be 16 millions of VPN instances on a router. In order to map MPLS labels to VPN instances, big labels are required. This document specifies the method to carry and distribute such big labels by piggybacking the big label mapping information for an IP route in the BGP Update message that is used to distribute the route itself.

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 http://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 January 02, 2014.

Copyright Notice

Copyright (c) 2013 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 (http://trustee.ietf.org/license-info) in effect on the date of 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.

This document may contain material from IETF Documents or IETF Contributions published or made publicly available before November 10, 2008. The person(s) controlling the copyright in some of this material may not have granted the IETF Trust the right to allow modifications of such material outside the IETF Standards Process. Without obtaining an adequate license from the person(s) controlling the copyright in such materials, this document may not be modified outside the IETF Standards Process, and derivative works of it may not be created outside the IETF Standards Process, except to format it for publication as an RFC or to translate it into languages other than English.

Table of Contents

• 1. Introduction
• 1.1. Requirement Language
• 1.2. Terminology
• 2. Motivation
• 3. Big Labels
• 4. AFI/SAFI
• 5. Big Label in NLRI
• 6. Capability Anouncement
• 7. IANA Considerations
• 8. Security Considerations
• 9. References
• 9.1. Normative References
• 9.2. Informative References
• Authors' Addresses

1. Introduction

Network virtualization and server virtualization are being designed and deployed in data center networks, and new data encapsulation methods and protocols are being defined and specified, for example, VXLAN, NVGRE and NVO3. The general idea is to add a new virtual network header so that a physical network can be used to support millions (16M) of virtualized overlaid networks. Network overlay virtualization have placed a new requirement on the access method to such virtulized networks.

BGP/MPLS IP VPNs of [RFC2547] and [RFC4364], provide a market-proven technology and solution for end-to-end IP VPNs. In BGP/MPLS IP VPNs, all the customer sites are connected to the service provider networks through PE-CE link. It is desirable to extend the BGP/MPLS scheme so that customers can access their virtualized networks hosted in a data center by using BGP/MPLS IP VPNs.

In the data plane of BGP/MPLS IP VPNs, the customer VPN/VRF instances are represented by an MPLS label (VPN label) locally assgined by the PE connecting to CE. Since MPLS labels are 20 bits long, a PE can maximally support 1 million VPNs/VRFs, but PE is required to support 16 millions of virtual networks that are being standardized in VXLAN, NVGRE and NVO3. When BGP/MPLS IP VPNs are extended to access virtualized networks in data centers, [I-D.draft-renwei-l3vpn-big-label] describes several use cases and solutions to use big labels to represent the VPN and maps them to virtual network instances.

The big label information mapped to VPN routes can be carried and distributed by BGP-4 in the framework of [RFC3107]. This document specifies the method to carry and distribute such big labels by piggybacking the big label mapping information for an IP route in the BGP Update message that is used to distribute the route itself.

1.1. Requirement Language

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119].

1.2. Terminology

The following terms are used in this document:

PE

Provider Edge, the provider edge router connected to CE.

CE

Customer Edge, the customer edge router connected to PE

MPLS LSR

MPLS label switch router

IANA

Internet Assigned Numbers Authority

AFI

Address Family Identifier

SAFI

Subsequent Address Family Identifier

2. Motivation

This document tries to make L3 VPN BGP signaling work with the MPLS big label specified in [I-D.draft-renwei-mpls-big-label]. With the proposed method, the MPLS big label can be carried in BGP NLRI and mapped to BGP routes.

The extension of BGP is used to connect a customer site to its virtualized network hosted in a data center by using, for example, VXLAN, NVGRE or NVO3.

Take NVO3 as an example. NVO3 is an on-going effort to standardize solutions to data center virtualizaiton with the goal of providing viable data encapsulation and protocols across a scaling range of a few thousand VMs to several million VMs running on greater than one hundred thousand physical servers. NVO3 considers approaches to multi-tenancy that reside at the network layer rather than using traditional isolation mechanisms that rely on the underlying layer 2 technology (e.g. VLANs).

Based on NVO3 framework and problem statement, NVO3 will deliver 16 million virtual networks in a physical data center.

                                   ..................................
                                   .                                .
                                   .                 +------------+ .
                                   .                 |+----+ +---+| .
                                   .                 ||    | |VM || .
                                   .                /+|NVE | +---+| .
                                   .               / ||    | +---+| .
-----+          +----------------+ . +------------+  ||    | |VM || .
     |          |                | . |            |  |+----+ +---+| .
 +---+--+   +------+         +---+---+--+ NVO3    |  |   Server   | .
 |  CE  |---|  PE  | MPLS    |  PE-NVE  | network |  +------------+ .
 |device|   |device| network |  device  | over L3 |                 .
 +---+--+   +------+         +---+---+--+ network |  +------------+ .
     |       /  |                | . |            |  |+----+ +---+| .
-----+      /   +----------------+ . +------------+  ||    | |VM || .
           /                       .               \ ||NVE | +---+| .
-----+    /                        .                \+|    | +---+| .
     |   /                         .                 ||    | |VM || .
 +---+--+                          .                 |+----+ +---+| .
 |  CE  |                          .                 |   Server   | .
 |device|                          . Data Center     +------------+ .
 +---+--+                          ..................................
     |
-----+

Potentially, PE-NVE needs to support 16 million of VRFs. The extension of BGP described in this document can be used between PE-NVE and PE for the purpose of association between L3VPN labels and network virtualizaiton instances in the NVO3 network.

3. Big Labels

A PE device uses VPN labels to find the associated VRFs for VPN packet forwarding. When L3VPN is used to access and connect a virtual network hosted in a data center by using VXLAN, NVGRE or NVO3, a label space having a minimal set of 16 million labels is required.

 0                   1                   2                   3
 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|            Big Label Indicator        | Exp |S|       TTL     |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|            Big Label Value                                    |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

The document [I-D.draft-renwei-mpls-big-label] specifies an encoding format by adding a new label value field to the common label as follows:

When an MPLS LSR receivs an MPLS packet, it reads out the MPLS label. If the MPLS label is a Big Label Indicator, it will use the subsequent 32-bit value as the MPLS label for the forwarding purpose.

In what follows, we will describe how BGP-4 carries and distributes such big labels for IP routes.

4. AFI/SAFI

In BGP-4, label mapping information is carried as part of the Network Layer Reachability Information (NLRI) in the Multiprotocol Extensions of BGP-4 [RFC2858].

The AFI indicates, as usual, the address family of the associated route. The fact that the NLRI carries an MPLS big label is indicated by using a SAFI value, to be requested to IANA. The SAFI value 4 is currently assigned for common labels. The currently unassigned range is 8-63. Before IANA assigns an official SAFI value for big labels, 8 is temporarily used as the SAFI value to indicate that it carries big labels.

5. Big Label in NLRI

The Network Layer Reachability Information carrying big labels is encoded as one or more triples of the form (length, label, prefix), whose fields are described as follows:

      +---------------------------+
      |   Length (1 octet)        |
      +---------------------------+
      |   Label (4 octets)        |
      +---------------------------+
      |   Prefix (variable)       |
      +---------------------------+

1. Length: The Length field indicates the length in bits of the address prefix plus the label.
2. Label: The Label field carries a 4-octet Big Label Value of the big label format specified in [I-D.draft-renwei-mpls-big-label]. Note that the Big Label Indicator is not carried in NLRI; instead, it will be assigned by IANA. In the data plane, when encoding a packet for forwarding, both the Big Label Indicator and Big Label Value must be encoded in the MPLS header as specified in [I-D.draft-renwei-mpls-big-label].
3. Prefix: The Prefix field contains address prefixes followed by enough trailing bits to make the end of the field fall on an octet boundary. Note that the value of trailing bits is irrelevant.

All rules and restrictions applicable to the SAFI value 4 is also applicable to the SAFI value 8 (subject to IANA) except for the Label field must be 4 octets. In particular, the following usage rules for SAFI value 4 also applies to the SAFI value for big labels:

The label(s) specified for a particular route (and associated with its address prefix) must be assigned by the LSR which is identified by the value of the Next Hop attribute of the route.

When a BGP speaker redistributes a route, the label(s) assigned to that route must not be changed (except by omission), unless the speaker changes the value of the Next Hop attribute of the route.

A BGP speaker can withdraw a previously advertised route (as well as the binding between this route and a label) by either (a) advertising a new route (and a label) with the same NLRI as the previously advertised route, or (b) listing the NLRI of the previously advertised route in the Withdrawn Routes field of an Update message. The label information carried (as part of NLRI) in the Withdrawn Routes field should be set to 0x800000. (Of course, terminating the BGP session also withdraws all the previously advertised routes.)

6. Capability Anouncement

A BGP speaker that uses MP-BGP [RFC2858] to carry label mapping information should use the Capabilities Optional Parameter, as defined in [RFC2842], to inform its peers about this capability. The MP_EXT Capability Code, as defined in [RFC2858], is used to advertise the (AFI, SAFI) pairs available on a particular connection.

A BGP speaker should not advertise this capability to another BGP speaker unless there is a Label Switched Path (LSP) between the two speakers.

7. IANA Considerations

The requirements on IANA are specified in other related documents [I-D.draft-renwei-mpls-big-label] and [I-D.draft-renwei-mpls-bgp-big-label], which request a reserved label to represent Big Label Indicator and BGP capabilities for big labels.

8. Security Considerations

This draft does not add any additional security implications to the BGP/MPLS IP VPNs. All existing authentication and security mechanisms for BGP and MPLS still apply.

9. References

9.1. Normative References

9.2. Informative References

Authors' Addresses