idnits 2.17.1 draft-ietf-pcn-baseline-encoding-03.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document seems to contain a disclaimer for pre-RFC5378 work, and may have content which was first submitted before 10 November 2008. The disclaimer is necessary when there are original authors that you have been unable to contact, or if some do not wish to grant the BCP78 rights to the IETF Trust. If you are able to get all authors (current and original) to grant those rights, you can and should remove the disclaimer; otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (April 7, 2009) is 5496 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-11) exists of draft-ietf-pcn-architecture-10 == Outdated reference: A later version (-07) exists of draft-ietf-tsvwg-admitted-realtime-dscp-05 == Outdated reference: A later version (-10) exists of draft-ietf-tsvwg-ecn-tunnel-02 Summary: 1 error (**), 0 flaws (~~), 4 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Congestion and Pre Congestion T. Moncaster 3 Internet-Draft BT 4 Intended status: Standards Track B. Briscoe 5 Expires: October 9, 2009 BT & UCL 6 M. Menth 7 University of Wuerzburg 8 April 7, 2009 10 Baseline Encoding and Transport of Pre-Congestion Information 11 draft-ietf-pcn-baseline-encoding-03 13 Status of This Memo 15 This Internet-Draft is submitted to IETF in full conformance with the 16 provisions of BCP 78 and BCP 79. This document may contain material 17 from IETF Documents or IETF Contributions published or made publicly 18 available before November 10, 2008. The person(s) controlling the 19 copyright in some of this material may not have granted the IETF 20 Trust the right to allow modifications of such material outside the 21 IETF Standards Process. Without obtaining an adequate license from 22 the person(s) controlling the copyright in such materials, this 23 document may not be modified outside the IETF Standards Process, and 24 derivative works of it may not be created outside the IETF Standards 25 Process, except to format it for publication as an RFC or to 26 translate it into languages other than English. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF), its areas, and its working groups. Note that 30 other groups may also distribute working documents as Internet- 31 Drafts. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 The list of current Internet-Drafts can be accessed at 39 http://www.ietf.org/ietf/1id-abstracts.txt. 41 The list of Internet-Draft Shadow Directories can be accessed at 42 http://www.ietf.org/shadow.html. 44 This Internet-Draft will expire on October 9, 2009. 46 Copyright Notice 48 Copyright (c) 2009 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents in effect on the date of 53 publication of this document (http://trustee.ietf.org/license-info). 54 Please review these documents carefully, as they describe your rights 55 and restrictions with respect to this document. 57 Abstract 59 The objective of Pre-Congestion Notification (PCN) is to protect the 60 quality of service (QoS) of inelastic flows within a Diffserv domain. 61 The overall rate of the PCN-traffic is metered on every link in the 62 PCN-domain, and PCN-packets are appropriately marked when certain 63 configured rates are exceeded. The level of marking allows the 64 boundary nodes to make decisions about whether t o admit or block a 65 new flow request, and (in abnormal circumstances) whether to 66 terminate some of the existing flows, thereby protecting the QoS of 67 previously admitted flows. This document specifies how such marks 68 are to be encoded into the IP header by re-using the ECN codepoints 69 within this controlled domain. The baseline encoding described here 70 provides for only two PCN encoding states, unmarked and marked. 72 Table of Contents 74 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 75 2. Requirements notation . . . . . . . . . . . . . . . . . . . . 5 76 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 77 4. Encoding two PCN States in IP . . . . . . . . . . . . . . . . 5 78 4.1. Valid and Invalid Codepoint Transitions . . . . . . . . . 6 79 4.2. Rationale for Encoding . . . . . . . . . . . . . . . . . . 7 80 4.3. PCN-Compatible DiffServ Codepoints . . . . . . . . . . . . 8 81 5. Rules for Experimental Encoding Schemes . . . . . . . . . . . 8 82 6. Backwards Compatibility . . . . . . . . . . . . . . . . . . . 8 83 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 84 8. Security Considerations . . . . . . . . . . . . . . . . . . . 9 85 9. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 9 86 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 9 87 11. Comments Solicited . . . . . . . . . . . . . . . . . . . . . . 10 88 12. References . . . . . . . . . . . . . . . . . . . . . . . . . . 10 89 12.1. Normative References . . . . . . . . . . . . . . . . . . . 10 90 12.2. Informative References . . . . . . . . . . . . . . . . . . 10 91 Appendix A. PCN Deployment Considerations . . . . . . . . . . . . 11 92 A.1. Choice of Suitable DSCPs . . . . . . . . . . . . . . . . . 11 93 A.2. Rationale for Using ECT(0) for Not Marked . . . . . . . . 11 95 1. Introduction 97 The objective of Pre-Congestion Notification (PCN) is to protect the 98 quality of service (QoS) of inelastic flows within a Diffserv domain, 99 in a simple, scalable and robust fashion. The overall rate of the 100 PCN-traffic is metered on every link in the PCN-domain, and PCN- 101 packets are appropriately marked when certain configured rates are 102 exceeded. These configured rates are below the rate of the link thus 103 providing notification before any congestion occurs (hence "pre- 104 congestion notification"). The level of marking allows the boundary 105 nodes to make decisions about whether to admit or block a new flow 106 request, and (in abnormal circumstances) whether to terminate some of 107 the existing flows, thereby protecting the QoS of previously admitted 108 flows. 110 This document specifies how these PCN marks are encoded into the IP 111 header by re-using the bits of the ECN field. It also describes how 112 packets are identified as belonging to a PCN flow. Some deployment 113 models require two PCN encoding states, others require more. The 114 baseline encoding described here only provides for two PCN encoding 115 states. However the encoding can be easily extended to provide more 116 states and rules for such extensions are given in this document. 118 Changes from previous drafts (to be removed by the RFC Editor): 120 From -02 to -03: 122 Extensive changes to address comments made by Gorry Fairhurst 123 including: 125 * Abstract re-written. 127 * Clarified throughout that this re-uses the ECN bits in the IP 128 header. 130 * Re-arranged order of terminology section for clarity. 132 * Table 2 replaced with new table and text. 134 * Security considerations re-written. 136 * Appendixes re-written to improve clarity. 138 * Numerous minor nits and language changes throughout. 140 Extensive other minor changes throughout. 142 From -01 to -02: 144 Removed Appendix A and replaced with reference to 145 [I-D.ietf-tsvwg-ecn-tunnel] 147 Moved Appendix B into main body of text. 149 Changed Appendix C to give deployment advice. 151 Minor changes throughout including checking consistency of 152 capitalisation of defined terms. 154 Clarified that LU was deliberately excluded from encoding. 156 From -00 to -01: 158 Added section on restrictions for extension encoding schemes. 160 Included table in Appendix showing encoding transitions at 161 different PCN nodes. 163 Checked for consistency of terminology. 165 Minor language changes for clarity. 167 Changes from previous filename 169 Filename changed from draft-moncaster-pcn-baseline-encoding. 171 Terminology changed for clarity (PCN-compatible DSCP and PCN- 172 enabled packet). 174 Minor changes throughout. 176 Modified meaning of ECT(1) state to EXP. 178 Moved text relevant to behaviour of nodes into appendix for later 179 transfer to new document on edge behaviours. 181 From draft-moncaster -01 to -02: 183 Minor changes throughout including tightening up language to 184 remain consistent with the PCN Architecture terminology 186 From draft-moncaster -00 to -01: 188 Change of title from "Encoding and Transport of (Pre-)Congestion 189 Information from within a DiffServ Domain to the Egress" 191 Extensive changes to Introduction and abstract. 193 Added a section on the implications of re-using a DSCP. 195 Added appendix listing possible operator scenarios for using this 196 baseline encoding. 198 Minor changes throughout. 200 2. Requirements notation 202 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 203 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 204 document are to be interpreted as described in [RFC2119]. 206 3. Terminology 208 The following terms are used in this document: 210 o PCN-compatible Diffserv codepoint - a Diffserv codepoint for which 211 the ECN field is used to carry PCN markings rather than [RFC3168] 212 markings. 214 o PCN-marked - codepoint indicating packets that have been marked at 215 a PCN-interior-node using some PCN marking behaviour 216 [I-D.ietf-pcn-marking-behaviour]. Abbreviated to PM. 218 o Not-marked - codepoint indicating packets that are PCN-capable, 219 but are not PCN-marked. Abbreviated to NM. 221 o PCN-enabled codepoints - collective term for all NM and PM 222 codepoints. By definition, packets carrying such codepoints are 223 PCN-packets. 225 o not-PCN - packets that are not PCN-enabled. 227 In addition, the document uses the terminology defined in 228 [I-D.ietf-pcn-architecture]. 230 4. Encoding two PCN States in IP 232 The PCN encoding states are defined using a combination of the DSCP 233 and ECN fields within the IP header. The baseline PCN encoding 234 closely follows the semantics of ECN [RFC3168]. It allows the 235 encoding of two PCN states: Not-marked and PCN-marked. It also 236 allows for traffic that is not PCN-capable to be marked as such (not- 237 PCN). Given the scarcity of codepoints within the IP header the 238 baseline encoding leaves one codepoint free for experimental use. 239 The following table defines how to encode these states in IP: 241 +---------------+-------------+-------------+-------------+---------+ 242 | ECN codepoint | Not-ECT | ECT(0) (10) | ECT(1) (01) | CE (11) | 243 | | (00) | | | | 244 +---------------+-------------+-------------+-------------+---------+ 245 | DSCP n | not-PCN | NM | EXP | PM | 246 +---------------+-------------+-------------+-------------+---------+ 248 Where DSCP n is a PCN-compatible DiffServ codepoint (see Section 4.3) 249 and EXP means available for Experimental use. N.B. we deliberately 250 reserve this codepoint for experimental use only (and not local use) 251 to prevent future compatability issues. 253 Table 1: Encoding PCN in IP 255 The following rules apply to all PCN traffic: 257 o PCN-traffic MUST be marked with a PCN-compatible DiffServ 258 Codepoint. To conserve DSCPs, DiffServ Codepoints SHOULD be 259 chosen that are already defined for use with admission controlled 260 traffic, such as the Voice-Admit codepoint defined in 261 [I-D.ietf-tsvwg-admitted-realtime-dscp]. Guidelines for mixing 262 traffic-types within a PCN-domain are given in 263 [I-D.ietf-pcn-marking-behaviour]. 265 o Any packet that is not-PCN but which shares the same DiffServ 266 codepoint as PCN-enabled traffic MUST have the ECN field equal to 267 00. 269 4.1. Valid and Invalid Codepoint Transitions 271 A PCN-ingress-node MUST set the Not-marked (10) codepoint on any 272 arriving packet that belongs to a PCN-flow. It MUST set the not-PCN 273 (00) codepoint on all other packets. 275 A PCN-interior-node MUST observe the rules for valid and invalid 276 codepoint transitions as set out in the following table. The precise 277 rules governing which valid transition to use are set out in 278 [I-D.ietf-pcn-marking-behaviour] 279 +-------------------------------------------------+ 280 | Codepoint Out | 281 +--------------+-------------+-----------+-----------+-----------+ 282 | Codepoint in | not-PCN(00) | NM(10) | EXP(01) | PM(11) | 283 +--------------+-------------+-----------+-----------+-----------+ 284 | not-PCN(00) | Valid | Not valid | Not valid | Not valid | 285 +--------------+-------------+-----------+-----------+-----------+ 286 | NM(10) | Not valid | Valid | Not valid | Valid | 287 +--------------+-------------+-----------+-----------+-----------+ 288 | EXP(01)* | Not valid | Not valid | Valid | Valid | 289 +--------------+-------------+-----------+-----------+-----------+ 290 | PM(11) | Not valid | Not valid | Not valid | Valid | 291 +--------------+-------------+-----------+-----------+-----------+ 292 * This SHOULD cause an alarm to be raised at a higher layer. The 293 packet MUST be treated as if it carried the NM codepoint. 295 Table 2: Valid and Invalid Codepoint Transitions for 296 PCN-packets at PCN-interior-nodes 298 A PCN-egress-node SHOULD set the not-PCN (00) codepoint on all 299 packets it forwards out of the PCN-domain. The only exception to 300 this is if the PCN-egress-node is certain that revealing other 301 codepoints outside the PCN-domain won't contravene the guidance given 302 in [RFC4774]. 304 4.2. Rationale for Encoding 306 The exact choice of encoding was dictated by the constraints imposed 307 by existing IETF RFCs, in particular [RFC3168], [RFC4301] and 308 [RFC4774]. One of the tightest constraints was the need for any PCN 309 encoding to survive being tunnelled through either an IP in IP tunnel 310 or an IPSec Tunnel. [I-D.ietf-tsvwg-ecn-tunnel] explains this in 311 more detail. The main effect of this constraint is that any PCN 312 marking has to carry the 11 codepoint in the ECN field since this is 313 the only codepoint that is guaranteeed to be copied down into the 314 inner header upon decapsulation. An additional constraint is the 315 need to minimise the use of DiffServ codepoints as there is a limited 316 supply of standards track codepoints remaining. Section 4.3 explains 317 how we have minimised this still further by reusing pre-existing 318 Diffserv codepoint(s) such that non-PCN traffic can still be 319 distinguished from PCN traffic. There are a number of factors that 320 were considered before deciding to set 10 as the NM state. These 321 included similarity to ECN, presence of tunnels within the domain, 322 leakage into and out of PCN-domain and incremental deployment (see 323 Appendix A.2). 325 The encoding scheme above seems to meet all these constraints and 326 ends up looking very similar to ECN. This is perhaps not surprising 327 given the similarity in architectural intent between PCN and ECN. 329 4.3. PCN-Compatible DiffServ Codepoints 331 Equipment complying with the baseline PCN encoding MUST allow PCN to 332 be enabled for certain Diffserv codepoints. This document defines 333 the term "PCN-compatible Diffserv codepoint" for such a DSCP. To be 334 clear, any packets with such a DSCP will be PCN enabled only if they 335 also have their ECN field set to indicate a codepoint other than not- 336 PCN. 338 Enabling PCN marking behaviour disables any other marking behaviour 339 (e.g. enabling PCN disables the default ECN marking behaviour 340 introduced in [RFC3168]). All traffic scheduling and conditioning 341 behaviours are discussed in [I-D.ietf-pcn-marking-behaviour]. This 342 ensures compliance with the BCP guidance set out in [RFC4774]. 344 5. Rules for Experimental Encoding Schemes 346 Any experimental encoding scheme MUST follow these rules to ensure 347 backward compatibility with this baseline scheme: 349 o The 00 codepoint in the ECN field SHALL indicate not-PCN and MUST 350 NOT be changed to any otehr codepoint within a PCN-domain. 351 Therefore an ingress node wishing to disable PCN marking for a 352 packet within a PCN-compatible DiffServ Codepoint MUST set the ECN 353 field to 00. 355 o The 11 codepoint in the ECN field SHALL indicate PCN-marked 356 (though this does not exclude the 01 Experimental codepoint from 357 carrying the same meaning). 359 o Once set, the 11 codepoint in the ECN field MUST NOT be changed to 360 any other codepoint. 362 o Any experimental scheme MUST include details of all valid and 363 invalid codepoint transitions at any PCN nodes. 365 o Any experimental scheme MUST NOT update the meaning of the 00 and 366 11 codepoints defined above. 368 6. Backwards Compatibility 370 BCP 124 [RFC4774] gives guidelines for specifying alternative 371 semantics for the ECN field. It sets out a number of factors to be 372 taken into consideration. It also suggests various techniques to 373 allow the co-existence of default ECN and alternative ECN semantics. 374 The baseline encoding specified in this document defines PCN- 375 compatible DiffServ codepoints as no longer supporting the default 376 ECN semantics. As such this document is compatible with BCP 124. It 377 should be noted that this baseline encoding effectively disables end- 378 to-end ECN except where mechanisms are put in place to tunnel such 379 traffic across the PCN-domain. 381 7. IANA Considerations 383 This document makes no request to IANA. 385 8. Security Considerations 387 PCN-marking only carries a meaning within the confines of a PCN- 388 domain. Packets wishing to be treated as belonging to a PCN-flow 389 must carry a PCN-Compatible DSCP and a PCN-Enabled ECN codepoint. 390 This encoding document is intended to stand independently of the 391 architecture used to determine whether specific packets are 392 authorised to be PCN-marked, which will be described in separate 393 documents on PCN edge-node behaviour. 395 This document assumes the PCN-domain to be entirely under the control 396 of a single operator, or a set of operators who trust each other. 397 However future extensions to PCN might include inter-domain versions 398 where trust cannot be assumed between domains. If such schemes are 399 proposed they must ensure that they can operate securely despite the 400 lack of trust but such considerations are beyond the scope of this 401 document. 403 9. Conclusions 405 This document defines the baseline PCN encoding utilising a 406 combination of a PCN-enabled DSCP and the ECN field in the IP header. 407 This baseline encoding allows the existence of two PCN encoding 408 states, not-Marked and PCN-marked. It also allows for the co- 409 existence of competing traffic within the same DSCP so long as that 410 traffic does not require ECN support within the PCN-domain. The 411 encoding scheme is conformant with [RFC4774]. 413 10. Acknowledgements 415 This document builds extensively on work done in the PCN working 416 group by Kwok Ho Chan, Georgios Karagiannis, Philip Eardley, Anna 417 Charny, Joe Babiarz and others. Thanks to Ruediger Geib and Gorry 418 Fairhurst for providing detailed comments on this document. 420 11. Comments Solicited 422 (To be removed by the RFC-Editor.) Comments and questions are 423 encouraged and very welcome. They can be addressed to the IETF 424 congestion and pre-congestion working group mailing list 425 , and/or to the authors. 427 12. References 429 12.1. Normative References 431 [I-D.ietf-pcn-marking-behaviour] Eardley, P., "Marking 432 behaviour of PCN-nodes", dra 433 ft-ietf-pcn-marking- 434 behaviour-02 (work in 435 progress), March 2009. 437 [RFC2119] Bradner, S., "Key words for 438 use in RFCs to Indicate 439 Requirement Levels", BCP 14, 440 RFC 2119, March 1997. 442 [RFC3168] Ramakrishnan, K., Floyd, S., 443 and D. Black, "The Addition 444 of Explicit Congestion 445 Notification (ECN) to IP", 446 RFC 3168, September 2001. 448 [RFC4774] Floyd, S., "Specifying 449 Alternate Semantics for the 450 Explicit Congestion 451 Notification (ECN) Field", 452 BCP 124, RFC 4774, 453 November 2006. 455 12.2. Informative References 457 [I-D.ietf-pcn-architecture] Eardley, P., "Pre-Congestion 458 Notification (PCN) 459 Architecture", draft-ietf- 460 pcn-architecture-10 (work in 461 progress), March 2009. 463 [I-D.ietf-tsvwg-admitted-realtime-dscp] Baker, F., Polk, J., and M. 464 Dolly, "DSCP for Capacity- 465 Admitted Traffic", draft- 466 ietf-tsvwg-admitted- 467 realtime-dscp-05 (work in 468 progress), November 2008. 470 [I-D.ietf-tsvwg-ecn-tunnel] Briscoe, B., "Tunnelling of 471 Explicit Congestion 472 Notification", draft-ietf- 473 tsvwg-ecn-tunnel-02 (work in 474 progress), March 2009. 476 [RFC4301] Kent, S. and K. Seo, 477 "Security Architecture for 478 the Internet Protocol", 479 RFC 4301, December 2005. 481 [RFC5127] Chan, K., Babiarz, J., and 482 F. Baker, "Aggregation of 483 DiffServ Service Classes", 484 RFC 5127, February 2008. 486 Appendix A. PCN Deployment Considerations 488 A.1. Choice of Suitable DSCPs 490 The PCN Working Group chose not to define a single DSCP for use with 491 PCN for several reasons. Firstly the PCN mechanism is applicable to 492 a variety of different traffic classes. Secondly standards track 493 DSCPs are in increasingly short supply. Thirdly PCN should be seen 494 as being essentially a marking behaviour similar to ECN but intended 495 for inelastic traffic. The choice of which DSCP is most suitable for 496 a given PCN-domain is dependant on the nature of the traffic entering 497 that domain and the link rates of all the links making up that 498 domain. In PCN-domains with uniformly high link rates, the 499 appropriate DSCPs would currently be those for the Real Time Traffic 500 Class [RFC5127]. If the PCN domain includes lower speed links it 501 would also be appropriate to use the DSCPs of the other traffic 502 classes that [I-D.ietf-tsvwg-admitted-realtime-dscp] defines for use 503 with admission control, such as the three video classes CS4, CS3 and 504 AF4 and the Admitted Telephony Class. 506 A.2. Rationale for Using ECT(0) for Not Marked 508 The choice of which ECT codepoint to use for the Not Marked state was 509 based on the following considerations: 511 o [RFC3168] full functionality tunnel within the PCN-domain: Either 512 ECT is safe. 514 o Leakage of traffic into PCN-domain: ECT(1) is slightly less likely 515 to occur so might be considered safer. 517 o Leakage of traffic out of PCN-domain: Either ECT is equally unsafe 518 (since this would incorrectly indicate the traffic was ECN-capable 519 outside the controlled PCN-domain). 521 o Incremental deployment: Either codepoint is suitable providing 522 that the codepoints are used consistently. 524 o Conceptual consistency with other schemes: ECT(0) is conceptually 525 consistent with [RFC3168]. 527 Overall this seemed to suggest ECT(0) was most appropriate to use. 529 Authors' Addresses 531 Toby Moncaster 532 BT 533 B54/70, Adastral Park 534 Martlesham Heath 535 Ipswich IP5 3RE 536 UK 538 Phone: +44 1473 648734 539 EMail: toby.moncaster@bt.com 541 Bob Briscoe 542 BT & UCL 543 B54/77, Adastral Park 544 Martlesham Heath 545 Ipswich IP5 3RE 546 UK 548 Phone: +44 1473 645196 549 EMail: bob.briscoe@bt.com 551 Michael Menth 552 University of Wuerzburg 553 room B206, Institute of Computer Science 554 Am Hubland 555 Wuerzburg D-97074 556 Germany 558 Phone: +49 931 888 6644 559 EMail: menth@informatik.uni-wuerzburg.de