7 Network Working Group R. Hinden
8 Request for Comments: 2374 Nokia
9 Obsoletes: 2073 M. O'Dell
10 Category: Standards Track UUNET
16 An IPv6 Aggregatable Global Unicast Address Format
20 This document specifies an Internet standards track protocol for the
21 Internet community, and requests discussion and suggestions for
22 improvements. Please refer to the current edition of the "Internet
23 Official Protocol Standards" (STD 1) for the standardization state
24 and status of this protocol. Distribution of this memo is unlimited.
28 Copyright (C) The Internet Society (1998). All Rights Reserved.
32 This document defines an IPv6 aggregatable global unicast address
33 format for use in the Internet. The address format defined in this
34 document is consistent with the IPv6 Protocol [IPV6] and the "IPv6
35 Addressing Architecture" [ARCH]. It is designed to facilitate
36 scalable Internet routing.
38 This documented replaces RFC 2073, "An IPv6 Provider-Based Unicast
39 Address Format". RFC 2073 will become historic. The Aggregatable
40 Global Unicast Address Format is an improvement over RFC 2073 in a
41 number of areas. The major changes include removal of the registry
42 bits because they are not needed for route aggregation, support of
43 EUI-64 based interface identifiers, support of provider and exchange
44 based aggregation, separation of public and site topology, and new
45 aggregation based terminology.
47 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
48 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
49 document are to be interpreted as described in [RFC 2119].
58 Hinden, et. al. Standards Track [Page 1]
60 RFC 2374 IPv6 Global Unicast Address Format July 1998
63 2.0 Overview of the IPv6 Address
65 IPv6 addresses are 128-bit identifiers for interfaces and sets of
66 interfaces. There are three types of addresses: Unicast, Anycast,
67 and Multicast. This document defines a specific type of Unicast
70 In this document, fields in addresses are given specific names, for
71 example "subnet". When this name is used with the term "ID" (for
72 "identifier") after the name (e.g., "subnet ID"), it refers to the
73 contents of the named field. When it is used with the term "prefix"
74 (e.g. "subnet prefix") it refers to all of the addressing bits to
75 the left of and including this field.
77 IPv6 unicast addresses are designed assuming that the Internet
78 routing system makes forwarding decisions based on a "longest prefix
79 match" algorithm on arbitrary bit boundaries and does not have any
80 knowledge of the internal structure of IPv6 addresses. The structure
81 in IPv6 addresses is for assignment and allocation. The only
82 exception to this is the distinction made between unicast and
85 The specific type of an IPv6 address is indicated by the leading bits
86 in the address. The variable-length field comprising these leading
87 bits is called the Format Prefix (FP).
89 This document defines an address format for the 001 (binary) Format
90 Prefix for Aggregatable Global Unicast addresses. The same address
91 format could be used for other Format Prefixes, as long as these
92 Format Prefixes also identify IPv6 unicast addresses. Only the "001"
93 Format Prefix is defined here.
95 3.0 IPv6 Aggregatable Global Unicast Address Format
97 This document defines an address format for the IPv6 aggregatable
98 global unicast address assignment. The authors believe that this
99 address format will be widely used for IPv6 nodes connected to the
100 Internet. This address format is designed to support both the
101 current provider-based aggregation and a new type of exchange-based
102 aggregation. The combination will allow efficient routing
103 aggregation for sites that connect directly to providers and for
104 sites that connect to exchanges. Sites will have the choice to
105 connect to either type of aggregation entity.
114 Hinden, et. al. Standards Track [Page 2]
116 RFC 2374 IPv6 Global Unicast Address Format July 1998
119 While this address format is designed to support exchange-based
120 aggregation (in addition to current provider-based aggregation) it is
121 not dependent on exchanges for it's overall route aggregation
122 properties. It will provide efficient route aggregation with only
123 provider-based aggregation.
125 Aggregatable addresses are organized into a three level hierarchy:
129 - Interface Identifier
131 Public topology is the collection of providers and exchanges who
132 provide public Internet transit services. Site topology is local to
133 a specific site or organization which does not provide public transit
134 service to nodes outside of the site. Interface identifiers identify
137 ______________ ______________
138 --+/ \+--------------+/ \+----------
139 ( P1 ) +----+ ( P3 ) +----+
140 +\______________/ | |----+\______________/+--| |--
142 | ______________ / | |-+ ______________ / | |--
143 +/ \+ +-+--+ \ / \+ +----+
145 --+\______________/ / \ \______________/
151 ( S.A ) ( S.B ) ( P5 ) ( P6 )( S.C )
152 \___/ \___/ \___/ \___/ \___/
156 ( S.D ) ( S.E ) ( S.F )
159 As shown in the figure above, the aggregatable address format is
160 designed to support long-haul providers (shown as P1, P2, P3, and
161 P4), exchanges (shown as X1 and X2), multiple levels of providers
162 (shown at P5 and P6), and subscribers (shown as S.x) Exchanges
163 (unlike current NAPs, FIXes, etc.) will allocate IPv6 addresses.
164 Organizations who connect to these exchanges will also subscribe
165 (directly, indirectly via the exchange, etc.) for long-haul service
166 from one or more long-haul providers. Doing so, they will achieve
170 Hinden, et. al. Standards Track [Page 3]
172 RFC 2374 IPv6 Global Unicast Address Format July 1998
175 addressing independence from long-haul transit providers. They will
176 be able to change long-haul providers without having to renumber
177 their organization. They can also be multihomed via the exchange to
178 more than one long-haul provider without having to have address
179 prefixes from each long-haul provider. Note that the mechanisms used
180 for this type of provider selection and portability are not discussed
183 3.1 Aggregatable Global Unicast Address Structure
185 The aggregatable global unicast address format is as follows:
187 | 3| 13 | 8 | 24 | 16 | 64 bits |
188 +--+-----+---+--------+--------+--------------------------------+
189 |FP| TLA |RES| NLA | SLA | Interface ID |
190 | | ID | | ID | ID | |
191 +--+-----+---+--------+--------+--------------------------------+
193 <--Public Topology---> Site
196 <------Interface Identifier----->
200 FP Format Prefix (001)
201 TLA ID Top-Level Aggregation Identifier
202 RES Reserved for future use
203 NLA ID Next-Level Aggregation Identifier
204 SLA ID Site-Level Aggregation Identifier
205 INTERFACE ID Interface Identifier
207 The following sections specify each part of the IPv6 Aggregatable
208 Global Unicast address format.
210 3.2 Top-Level Aggregation ID
212 Top-Level Aggregation Identifiers (TLA ID) are the top level in the
213 routing hierarchy. Default-free routers must have a routing table
214 entry for every active TLA ID and will probably have additional
215 entries providing routing information for the TLA ID in which they
216 are located. They may have additional entries in order to optimize
217 routing for their specific topology, but the routing topology at all
218 levels must be designed to minimize the number of additional entries
219 fed into the default free routing tables.
226 Hinden, et. al. Standards Track [Page 4]
228 RFC 2374 IPv6 Global Unicast Address Format July 1998
231 This addressing format supports 8,192 (2^13) TLA ID's. Additional
232 TLA ID's may be added by either growing the TLA field to the right
233 into the reserved field or by using this format for additional format
236 The issues relating to TLA ID assignment are beyond the scope of this
237 document. They will be described in a document under preparation.
241 The Reserved field is reserved for future use and must be set to
244 The Reserved field allows for future growth of the TLA and NLA fields
245 as appropriate. See section 4.0 for a discussion.
247 3.4 Next-Level Aggregation Identifier
249 Next-Level Aggregation Identifier's are used by organizations
250 assigned a TLA ID to create an addressing hierarchy and to identify
251 sites. The organization can assign the top part of the NLA ID in a
252 manner to create an addressing hierarchy appropriate to its network.
253 It can use the remainder of the bits in the field to identify sites
254 it wishes to serve. This is shown as follows:
256 | n | 24-n bits | 16 | 64 bits |
257 +-----+--------------------+--------+-----------------+
258 |NLA1 | Site ID | SLA ID | Interface ID |
259 +-----+--------------------+--------+-----------------+
261 Each organization assigned a TLA ID receives 24 bits of NLA ID space.
262 This NLA ID space allows each organization to provide service to
263 approximately as many organizations as the current IPv4 Internet can
264 support total networks.
266 Organizations assigned TLA ID's may also support NLA ID's in their
267 own Site ID space. This allows the organization assigned a TLA ID to
268 provide service to organizations providing public transit service and
269 to organizations who do not provide public transit service. These
270 organizations receiving an NLA ID may also choose to use their Site
271 ID space to support other NLA ID's. This is shown as follows:
282 Hinden, et. al. Standards Track [Page 5]
284 RFC 2374 IPv6 Global Unicast Address Format July 1998
287 | n | 24-n bits | 16 | 64 bits |
288 +-----+--------------------+--------+-----------------+
289 |NLA1 | Site ID | SLA ID | Interface ID |
290 +-----+--------------------+--------+-----------------+
292 | m | 24-n-m | 16 | 64 bits |
293 +-----+--------------+--------+-----------------+
294 |NLA2 | Site ID | SLA ID | Interface ID |
295 +-----+--------------+--------+-----------------+
297 | o |24-n-m-o| 16 | 64 bits |
298 +-----+--------+--------+-----------------+
299 |NLA3 | Site ID| SLA ID | Interface ID |
300 +-----+--------+--------+-----------------+
302 The design of the bit layout of the NLA ID space for a specific TLA
303 ID is left to the organization responsible for that TLA ID. Likewise
304 the design of the bit layout of the next level NLA ID is the
305 responsibility of the previous level NLA ID. It is recommended that
306 organizations assigning NLA address space use "slow start" allocation
307 procedures similar to [RFC2050].
309 The design of an NLA ID allocation plan is a tradeoff between routing
310 aggregation efficiency and flexibility. Creating hierarchies allows
311 for greater amount of aggregation and results in smaller routing
312 tables. Flat NLA ID assignment provides for easier allocation and
313 attachment flexibility, but results in larger routing tables.
315 3.5 Site-Level Aggregation Identifier
317 The SLA ID field is used by an individual organization to create its
318 own local addressing hierarchy and to identify subnets. This is
319 analogous to subnets in IPv4 except that each organization has a much
320 greater number of subnets. The 16 bit SLA ID field support 65,535
323 Organizations may choose to either route their SLA ID "flat" (e.g.,
324 not create any logical relationship between the SLA identifiers that
325 results in larger routing tables), or to create a two or more level
326 hierarchy (that results in smaller routing tables) in the SLA ID
327 field. The latter is shown as follows:
338 Hinden, et. al. Standards Track [Page 6]
340 RFC 2374 IPv6 Global Unicast Address Format July 1998
343 | n | 16-n | 64 bits |
344 +-----+------------+-------------------------------------+
345 |SLA1 | Subnet | Interface ID |
346 +-----+------------+-------------------------------------+
348 | m |16-n-m | 64 bits |
349 +----+-------+-------------------------------------+
350 |SLA2|Subnet | Interface ID |
351 +----+-------+-------------------------------------+
353 The approach chosen for structuring an SLA ID field is the
354 responsibility of the individual organization.
356 The number of subnets supported in this address format should be
357 sufficient for all but the largest of organizations. Organizations
358 which need additional subnets can arrange with the organization they
359 are obtaining Internet service from to obtain additional site
360 identifiers and use this to create additional subnets.
364 Interface identifiers are used to identify interfaces on a link.
365 They are required to be unique on that link. They may also be unique
366 over a broader scope. In many cases an interfaces identifier will be
367 the same or be based on the interface's link-layer address.
368 Interface IDs used in the aggregatable global unicast address format
369 are required to be 64 bits long and to be constructed in IEEE EUI-64
370 format [EUI-64]. These identifiers may have global scope when a
371 global token (e.g., IEEE 48bit MAC) is available or may have local
372 scope where a global token is not available (e.g., serial links,
373 tunnel end-points, etc.). The "u" bit (universal/local bit in IEEE
374 EUI-64 terminology) in the EUI-64 identifier must be set correctly,
375 as defined in [ARCH], to indicate global or local scope.
377 The procedures for creating EUI-64 based Interface Identifiers is
378 defined in [ARCH]. The details on forming interface identifiers is
379 defined in the appropriate "IPv6 over <link>" specification such as
380 "IPv6 over Ethernet" [ETHER], "IPv6 over FDDI" [FDDI], etc.
382 4.0 Technical Motivation
384 The design choices for the size of the fields in the aggregatable
385 address format were based on the need to meet a number of technical
386 requirements. These are described in the following paragraphs.
388 The size of the Top-Level Aggregation Identifier is 13 bits. This
389 allows for 8,192 TLA ID's. This size was chosen to insure that the
390 default-free routing table in top level routers in the Internet is
394 Hinden, et. al. Standards Track [Page 7]
396 RFC 2374 IPv6 Global Unicast Address Format July 1998
399 kept within the limits, with a reasonable margin, of the current
400 routing technology. The margin is important because default-free
401 routers will also carry a significant number of longer (i.e., more-
402 specific) prefixes for optimizing paths internal to a TLA and between
405 The important issue is not only the size of the default-free routing
406 table, but the complexity of the topology that determines the number
407 of copies of the default-free routes that a router must examine while
408 computing a forwarding table. Current practice with IPv4 it is
409 common to see a prefix announced fifteen times via different paths.
411 The complexity of Internet topology is very likely to increase in the
412 future. It is important that IPv6 default-free routing support
413 additional complexity as well as a considerably larger internet.
415 It should be noted for comparison that at the time of this writing
416 (spring, 1998) the IPv4 default-free routing table contains
417 approximately 50,000 prefixes. While this shows that it is possible
418 to support more routes than 8,192 it is matter of debate if the
419 number of prefixes supported today in IPv4 is already too high for
420 current routing technology. There are serious issues of route
421 stability as well as cases of providers not supporting all top level
422 prefixes. The technical requirement was to pick a TLA ID size that
423 was below, with a reasonable margin, what was being done with IPv4.
425 The choice of 13 bits for the TLA field was an engineering
426 compromise. Fewer bits would have been too small by not supporting
427 enough top level organizations. More bits would have exceeded what
428 can be reasonably accommodated, with a reasonable margin, with
429 current routing technology in order to deal with the issues described
430 in the previous paragraphs.
432 If in the future, routing technology improves to support a larger
433 number of top level routes in the default-free routing tables there
434 are two choices on how to increase the number TLA identifiers. The
435 first is to expand the TLA ID field into the reserved field. This
436 would increase the number of TLA ID's to approximately 2 million.
437 The second approach is to allocate another format prefix (FP) for use
438 with this address format. Either or a combination of these
439 approaches allows the number of TLA ID's to increase significantly.
441 The size of the Reserved field is 8 bits. This size was chosen to
442 allow significant growth of either the TLA ID and/or the NLA ID
445 The size of the Next-Level Aggregation Identifier field is 24 bits.
450 Hinden, et. al. Standards Track [Page 8]
452 RFC 2374 IPv6 Global Unicast Address Format July 1998
455 This allows for approximately sixteen million NLA ID's if used in a
456 flat manner. Used hierarchically it allows for a complexity roughly
457 equivalent to the IPv4 address space (assuming an average network
458 size of 254 interfaces). If in the future additional room for
459 complexity is needed in the NLA ID, this may be accommodated by
460 extending the NLA ID into the Reserved field.
462 The size of the Site-Level Aggregation Identifier field is 16 bits.
463 This supports 65,535 individual subnets per site. The design goal
464 for the size of this field was to be sufficient for all but the
465 largest of organizations. Organizations which need additional
466 subnets can arrange with the organization they are obtaining Internet
467 service from to obtain additional site identifiers and use this to
468 create additional subnets.
470 The Site-Level Aggregation Identifier field was given a fixed size in
471 order to force the length of all prefixes identifying a particular
472 site to be the same length (i.e., 48 bits). This facilitates
473 movement of sites in the topology (e.g., changing service providers
474 and multi-homing to multiple service providers).
476 The Interface ID Interface Identifier field is 64 bits. This size
477 was chosen to meet the requirement specified in [ARCH] to support
478 EUI-64 based Interface Identifiers.
482 The authors would like to express our thanks to Thomas Narten, Bob
483 Fink, Matt Crawford, Allison Mankin, Jim Bound, Christian Huitema,
484 Scott Bradner, Brian Carpenter, John Stewart, and Daniel Karrenberg
485 for their review and constructive comments.
489 [ALLOC] IAB and IESG, "IPv6 Address Allocation Management",
490 RFC 1881, December 1995.
492 [ARCH] Hinden, R., "IP Version 6 Addressing Architecture",
495 [AUTH] Atkinson, R., "IP Authentication Header", RFC 1826, August
498 [AUTO] Thompson, S., and T. Narten., "IPv6 Stateless Address
499 Autoconfiguration", RFC 1971, August 1996.
501 [ETHER] Crawford, M., "Transmission of IPv6 Packets over Ethernet
502 Networks", Work in Progress.
506 Hinden, et. al. Standards Track [Page 9]
508 RFC 2374 IPv6 Global Unicast Address Format July 1998
511 [EUI64] IEEE, "Guidelines for 64-bit Global Identifier (EUI-64)
512 Registration Authority",
513 http://standards.ieee.org/db/oui/tutorials/EUI64.html,
516 [FDDI] Crawford, M., "Transmission of IPv6 Packets over FDDI
517 Networks", Work in Progress.
519 [IPV6] Deering, S., and R. Hinden, "Internet Protocol, Version 6
520 (IPv6) Specification", RFC 1883, December 1995.
522 [RFC2050] Hubbard, K., Kosters, M., Conrad, D., Karrenberg, D.,
523 and J. Postel, "Internet Registry IP Allocation
524 Guidelines", BCP 12, RFC 1466, November 1996.
526 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
527 Requirement Levels", BCP 14, RFC 2119, March 1997.
529 7.0 Security Considerations
531 IPv6 addressing documents do not have any direct impact on Internet
532 infrastructure security. Authentication of IPv6 packets is defined
562 Hinden, et. al. Standards Track [Page 10]
564 RFC 2374 IPv6 Global Unicast Address Format July 1998
567 8.0 Authors' Addresses
575 Phone: 1 408 990-2004
576 EMail: hinden@iprg.nokia.com
580 UUNET Technologies, Inc.
585 Phone: 1 703 206-5890
586 EMail: mo@uunet.uu.net
591 170 West Tasman Drive
592 San Jose, CA 95134-1706
595 Phone: 1 408 527-8213
596 EMail: deering@cisco.com
618 Hinden, et. al. Standards Track [Page 11]
620 RFC 2374 IPv6 Global Unicast Address Format July 1998
623 9.0 Full Copyright Statement
625 Copyright (C) The Internet Society (1998). All Rights Reserved.
627 This document and translations of it may be copied and furnished to
628 others, and derivative works that comment on or otherwise explain it
629 or assist in its implementation may be prepared, copied, published
630 and distributed, in whole or in part, without restriction of any
631 kind, provided that the above copyright notice and this paragraph are
632 included on all such copies and derivative works. However, this
633 document itself may not be modified in any way, such as by removing
634 the copyright notice or references to the Internet Society or other
635 Internet organizations, except as needed for the purpose of
636 developing Internet standards in which case the procedures for
637 copyrights defined in the Internet Standards process must be
638 followed, or as required to translate it into languages other than
641 The limited permissions granted above are perpetual and will not be
642 revoked by the Internet Society or its successors or assigns.
644 This document and the information contained herein is provided on an
645 "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
646 TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
647 BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
648 HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
649 MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
674 Hinden, et. al. Standards Track [Page 12]