RFC 1467

Network Working Group                                        C. Topolcic
Request for Comments: 1467                                          CNRI
Obsoletes: 1367                                              August 1993

               Status of CIDR Deployment in the Internet

Status of this Memo

   This memo provides information for the Internet community.  It does
   not specify an Internet standard.  Distribution of this memo is


   This document describes the current status of the development and
   deployment of CIDR technology into the Internet. This document
   replaces RFC 1367, which was a schedule for the deployment of IP
   address space management procedures to support route aggregation.
   Since all the milestones proposed in RFC 1367 except for the delivery
   and installation of CIDR software were met, it does not seem
   appropriate to issue an updated schedule. Rather, this document is
   intended to provide information about how this effort is proceeding,
   which may be of interest to the community.

1. Background

   The Internet's exponential growth has led to a number of difficulties
   relating to the management of IP network numbers.  The administrative
   overhead of allocating ever increasing volumes of IP network numbers
   for global users has stressed the organizations that perform this
   function.  The volume of IP network numbers that are reachable
   through the Internet has taxed a number of routers' ability to manage
   their forwarding tables.  The poor utilization of allocated IP
   network numbers has threatened to deplete the Class A and Class B
   address space.

   During the past few years, a consensus has emerged among the Internet
   community in favor of a number of mechanisms to relieve these
   problems for the mid-term.  These mechanisms are expected to be put
   into place in the short term and to provide relief for the mid-term.
   Fundamental changes to the Internet protocols to ensure the
   Internet's continued long term growth and well being are being
   explored and are expected to succeed the mid-term mechanisms.

   The global Internet community have been cooperating closely in such
   forums as the IETF and its working groups, the IEPG, the NSF Regional
   Techs Meetings, INET, INTEROP, FNC, FEPG, and other assemblies in

Topolcic                                                        [Page 1]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

   order to ensure the continued stable operation of the Internet.
   Recognizing the need for the mid-term mechanisms and receiving
   support from the Internet community, the US Federal Agencies proposed
   procedures to assist the deployment of these mid-term mechanisms.
   These procedures were originally described in RFC 1366 [1], which was
   recently made obsolete by RFC 1466 [2].  In October 1992, a schedule
   was proposed for the implementation of the procedures, described in
   RFC 1367 [3].

2. Milestones that have been met

   Most of the milestones of the proposed schedule were implemented on
   time. These milestones are shown below, essentially as they appear in
   [3], but with further comment where appropriate:

      1) 31 October 92:

         The following address allocation procedures were continued:

         a) Initial set of criteria for selecting regional address
            registries were put into place, and requests from
            prospective regional registries were accepted by the

            The Reseaux IP Europeens Network Coordination Centre
            (RIPE NCC) requested to become a regional registry.
            As per the addressing plan of RFC 1366, the RIPE NCC
            was given the block to to
            administer for the European Internet community. The RIPE
            NCC had previously and independently obtained the block
   to Although this block had been
            allocated before RFC 1366, the RIPE NCC was able to manage
            it according to the guidelines in RFC 1366.

         b) Class A network numbers were put on reserve for possible
            future use. The unreserved Class A numbers became very
            difficult to obtain.

         c) Class B network numbers were issued only when
            reasonably justified.  Whenever possible, a block of C's
            was issued rather than a B. The requirements for
            allocating a Class B became progressively more constrained
            until the date in step (3).

Topolcic                                                        [Page 2]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

         d) Class C network numbers were allocated according to the
            addressing plan of [1], now obsoleted by [2].  Allocation
            continued to be performed by the Internet Registry (IR)
            for regions of the world where an appropriate regional
            registry had not yet been designated by the IANA.

      2) 14 February 93:

         The schedule in [3] was re-evaluated, and there appeared to
         be no reason to readjust it, so it was continued as
         originally set out.

      3) 15 April 93:

         a) The IR began to allocate all networks according to the
            addressing plan of [1], now obsoleted by [2], in
            appropriately sized blocks of Class C numbers.

         b) Class B network numbers became difficult to obtain,
            following the recommendation of the addressing plan and
            were only issued when justified.

   Furthermore, throughout this time period, network service providers
   have requested blocks of network numbers from the Class C address
   space for the purpose of further allocating them to their clients.
   The network service providers were allocated such space by the RIPE
   NCC or the IR, acting for North America and the Pacific Rim. This
   process has started to distribute the function of address
   registration to a more regional level, closer to the end users. The
   process has operated as hoped for, with no major problems.

3. Milestone that has not been met

   The proposed schedule of [3] stated that 6 June 1993 was the date
   when an address aggregation mechanism would be generally available in
   the Internet. Although this target date was based on the plans as
   stated by the router vendors and was reasonable at the time the
   schedule in [3] was formulated, it has slipped.  Nevertheless, the
   continuation of that schedule has so far not added significantly to
   the problems of the Internet. The rest of this document looks at the
   current situation and what can be expected in the near future.

4. Current status of address aggregation mechanisms in commercial

   Although RFCs 1366, 1466, and 1367 do not depend on any specific
   address aggregation technology, there is consensus in the Internet
   community to use Classless Inter-Domain Routing (CIDR) [4]. CIDR is

Topolcic                                                        [Page 3]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

   supported by BGP-4 and IDRP. Most router vendors are working on BGP-
   4, first, and there is a consensus to use BGP-4 to support the
   initial deployment of CIDR in the Internet.

   The following paragraphs describe the implementation status and plans
   of software to support CIDR in various router vendors' products,
   listed in alphabetical order.  Some speculation is necessarily
   involved in deriving these projections.  See also the minutes of the
   July 1993 meeting of the BGP Deployment Working Group of the IETF

   3Com's BGP-4 code has been tested internally. They have code that
   accepts, forwards and manages aggregated routes properly, and they
   are ready to test it for interoperability with other vendors. They
   have yet to implement the code that forms the route aggregates. They
   expect to have Beta code done by September, and full release code
   shortly thereafter. The initial implementation will not support de-
   aggregation.  Their plans here are not yet formulated. They will
   support de-aggregation if necessary.

   ANS has a BGP-4 implementation that is being tested internally.  It
   is stable enough to begin testing for interoperability with other
   vendors' implementations.  Depending of the results of
   interoperability testing, this code could be deployed into the ANSNET
   by August.  This delay is primarily because some routers are running
   older code, and they all need to be upgraded to GATED before they can
   all support BGP-4 internally. So the ability to support CIDR looks
   like it is about one to two months away. This code will not support
   controlled de-aggregation, but de-aggregation will be supported if

   BBN plans to complete it's development of BGP-4 by early Summer 1994.
   Initial plans are to implement both aggregation and controlled de-
   aggregation with an early release of the software.

   Cisco's BGP-4 implementation is under development at this time.
   There is pre-Beta code available for people to begin testing.  It is
   expected that the code will be stable sometime during the summer of
   1993 and will be made available for limited deployment at that time.
   This BGP-4 code will implement aggregation. It will not be part of
   the normal release cycle at this time.  It will be available in a
   special software release based on the 9.21 release. This initial
   BGP-4 code will not implement controlled de-aggregation, but Cisco
   plans on implementing de-aggregation.

   Proteon's BGP-4 code has been tested internally. They are ready to
   test it for interoperability with other vendors. If this works out
   reasonably well, then it is reasonable to expect that they can start

Topolcic                                                        [Page 4]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

   to deploy this as Beta code by August, with a target of full release
   in the fall. This initial implementation will not support aggregation
   or de-aggregation. Aggregation will be implemented soon thereafter,
   but their plans for de-aggregation are not yet formulated.  They will
   implement de-aggregation if necessary.

   Wellfleet is aiming at having beta code implementing BGP-4 roughly in
   early 1994. This code will include controlled de-aggregation.

5. Rate of growth

   MERIT periodically publishes the number of networks in the
   NSFNET/ANSNET policy routing database.  Analysis of this data
   suggests that the number of entries in this database is growing at
   approximately 8% per month, or doubling every nine or ten months [6].

   Although there are currently over 13K networks in the NSFNET/ANSNET
   policy routing database, a number of them are not active. That is,
   they are not announced to the NSFNET/ANSNET Backbone. The 10K active
   network point was passed in late June. Assuming that the number of
   active networks continues to grow at the same rate as in the past, it
   can be projected that the 12K active network point will be reached
   sometime in approximately late September 1993 and that the 25K active
   network point will be reached sometime in mid-94 (two high water
   marks whose relevance will become apparent below).

   The NSFNET/ANSNET routing database includes only those networks that
   meet the NSF Acceptable Use Policy (AUP) or the ANSNET CO+RE AUP.
   There are a number of networks connected to the Internet that do not
   meet these criteria. Although they are not in the NSFNET/ANSNET
   routing database, they are in the forwarding tables of a number of
   network providers. Currently, the number of networks that are
   connected to other known service providers but are not in the
   NSFNET/ANSNET routing database is significantly smaller than (less
   than 25% of) the number that are in the NSFNET/ANSNET database. There
   is no estimate available for the rate of growth of the number of such
   non-NSFNET/ANSNET networks. It is assumed here that the growth rate
   of these networks is approximately the same as that of AUP networks
   in the NSFNET/ANSNET routing database.

   Analysis of the more than 13K networks in the NSFNET/ANSNET routing
   database, as well as the allocated but unconnected networks, suggests
   that CIDR deployment should have a significant impact on the number
   of forwarding table entries that any router needs to maintain, and
   its rate of growth.  However, an in-depth study was begun at the July
   1993 meeting of the BGP Deployment Working Group of the IETF [5] to
   (among other goals) evaluate the impact of CIDR on the growth rate of
   router forwarding tables.

Topolcic                                                        [Page 5]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

6. Capacity of deployed networks

   The following paragraphs describe the current occupancy of the
   forwarding tables of the routers of several transit network providers
   and their expected capacities and an estimate of the time when that
   capacity would be reached if the growth rate were to continue as
   today. This list is a subset of all relevant providers, but is
   considered approximately representative of the situation of other
   network providers. It is shown in alphabetical order.

   ALTERNET nodes are Cisco routers, and currently carry approximately
   11K to 12K routes, both AUP and non-AUP. With their current
   configuration, they have enough memory so that they are expected to
   support up to approximately 35K routes.  If the rate at which the
   number of these routes is expected to grow is approximately the same
   as the rate that the NSFNET/ANSNET policy routing database is
   growing, then this number may be reached in late 1994.  However, if
   the growth rate continues unchecked, it is expected that the
   processing capacity of the routers will be surpassed before their
   memory is exhausted. It is expected that CIDR will be in place long
   before this point is reached.

   All ANSNET routers have recently been upgraded to AIX 3.2. This
   version supports up to 12K networks.  These routers currently carry
   only the active networks in the NSFNET/ANSNET routing database.  It
   is anticipated that the next version of router code will be deployed
   before September 1993, the projected date for when there will be 12K
   active networks.  This version will support 25K active networks.
   Although there are no current plans for a version of router code that
   supports more than 25K networks, it is believed that CIDR will help
   this situation.

   EBONE nodes are Cisco routers. They currently carry approximately 10K
   to 11K routes. With their current configuration, they may be able to
   support approximately 40K routes. However, the number of paths may be
   very relevant. The memory required for the BGP table (rather than the
   forwarding table) is a function of the number of paths.  If a new
   transatlantic link were to be added, EBONE could receive all the
   North American routes through it. This would add a new set of paths.
   Each such transatlantic link would increase the memory required by
   approximately 20%. Due to the network topology between North America
   and Europe, new transatlantic links tend to result in new paths, and
   therefore significant memory requirements. It is very difficult to
   predict the addition of future transatlantic links because they
   result from business or political requirements, not bandwidth

Topolcic                                                        [Page 6]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

   ESNET uses Cisco routers. However, it is already in trouble, but not
   because of the size of the forwarding tables. The problem is its need
   to maintain considerable configuration information describing which
   networks it should or should not accept from its neighbors, and the
   fact that this information must be stored in a non-volatile memory of
   limited size. CIDR aggregation is expected to help this problem.
   Also, ESNET plans to deploy BGP-4 and CIDR only after it is in a full
   release, so does not plan to participate in the initial BGP-4
   deployment. ESNET will upgrade their nodes to Cisco CSC-4's in the

   All SPRINTLINK and ICM nodes have recently been upgraded to Cisco
   CSC-4 routers with 16MB of memory. They will carry full routing,
   including not only the routes that the NSFNET/ANSNET carries, but
   also routes to networks that do not comply with the NSF or CO+RE
   AUPs. The SPRINT routers currently carry approximately 11K to 12K
   routes, and it is expected that they will be able to support up to
   approximately 25K routes, as currently configured. The 25K announced
   network point may be reached in approximately mid-1994. Again, it is
   expected that CIDR deployment will have a significant impact on this
   growth rate, well before this time.

7. Acknowledgements

   This report contains information from a number of sources, including
   vendors, operators, researchers, and organizations that foster
   cooperation in the Internet community. Specific organizations include
   the Intercontinental Engineering and Planning Group (IEPG), the BGP-4
   Deployment Working Group of the IETF, the Federal Networking Council
   (FNC), and the FNC Engineering and Planning Group (FEPG). Specific
   individuals include, in alphabetical order, Arun Arunkumar, Tony
   Bates, Mary Byrne, Bob Collet, Mike Craren, Dennis Ferguson, Tony
   Hain, Elise Gerich, Mark Knopper, John Krawczyk, Tony Li, Peter
   Lothberg, Andrew Partan, Gary Rucinski, Frank Solensky, and Jessica
   Yu. This report would not have been possible without the willingness
   of these people to make their information public for the good of the

8. References

   [1] Gerich, E., "Guidelines for Management of IP Address Space",
       RFC 1366, Merit, October 1992.

   [2] Gerich, E., "Guidelines for Management of IP Address Space",
       RFC 1466, Merit, May 1993.

   [3] Topolcic, C., "Schedule for IP Address Space Management
       Guidelines", RFC 1367, CNRI, October 1992.

Topolcic                                                        [Page 7]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

   [4] Fuller, V. et al, "Classless Inter-Domain Routing (CIDR): an
       Address Assignment and Aggregation Strategy", working draft
       obsoleting RFC 1338, BARRNet, February 1993.

   [5] Yu, J., "Minutes of the BGP Deployment Working Group
       (BGPDEPL)", MERIT, July 1993.

   [6] Solensky, F., Internet Growth Charts, "big-internet" mailing
       list, munnari.oz.au:big-internet/nsf-netnumbers-<yymm>.ps

9. Other relevant documents

       Huitema, C., "IAB Recommendation for an Intermediate Strategy
       to Address the Issue of Scaling", RFC 1481, Internet
       Architecture Board, July 1993.

       Knopper, M., "Minutes of the NSFNET Regional Techs Meeting",
       working draft, MERIT, June 1993.

       Knopper, M., and Richardson, S., " Aggregation Support in the
       NSFNET Policy-Based Routing Database", RFC 1482, MERIT, June

       Topolcic, C., "Notes of BGP-4/CIDR Coordination Meeting of 11
       March 93", working draft, CNRI, March 1993.

       Rekhter, Y., and Topolcic, C., "Exchanging Routing Information
       Across Provider/Subscriber Boundaries in the CIDR Environment",
       working draft, IBM Corp., CNRI, April 1993.

       Rekhter, Y., and Li, T., "An Architecture for IP Address
       Allocation with CIDR", working draft, IBM Corp., cisco Systems,
       February 1993.

       Gross, P., and P. Almquist, "IESG Deliberations on Routing and
       Addressing", RFC 1380, IESG, November 1992.

Topolcic                                                        [Page 8]

RFC 1467       Status of CIDR Deployment in the Internet     August 1993

10. Security Considerations

   Security issues are not discussed in this memo.

11. Author's Address

   Claudio Topolcic
   Corporation for National Research Initiatives
   895 Preston White Drive, Suite 100
   Reston, VA  22091

   Phone: (703) 620-8990
   EMail: topolcic@CNRI.Reston.VA.US

Topolcic                                                        [Page 9]