idnits 2.17.1 draft-ietf-avt-hc-mpls-reqs-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.5 on line 425. ** The document seems to lack an RFC 3978 Section 5.1 IPR Disclosure Acknowledgement -- however, there's a paragraph with a matching beginning. Boilerplate error? ** This document has an original RFC 3978 Section 5.4 Copyright Line, instead of the newer IETF Trust Copyright according to RFC 4748. ** The document seems to lack an RFC 3978 Section 5.4 Reference to BCP 78 -- however, there's a paragraph with a matching beginning. Boilerplate error? ** This document has an original RFC 3978 Section 5.5 Disclaimer, instead of the newer disclaimer which includes the IETF Trust according to RFC 4748. ** The document seems to lack an RFC 3979 Section 5, para. 1 IPR Disclosure Acknowledgement. ** The document seems to lack an RFC 3979 Section 5, para. 2 IPR Disclosure Acknowledgement. ** The document seems to lack an RFC 3979 Section 5, para. 3 IPR Disclosure Invitation. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** Missing expiration date. The document expiration date should appear on the first and last page. ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity -- however, there's a paragraph with a matching beginning. Boilerplate error? ** The document seems to lack a 1id_guidelines paragraph about the list of current Internet-Drafts -- however, there's a paragraph with a matching beginning. Boilerplate error? == The page length should not exceed 58 lines per page, but there was 6 longer pages, the longest (page 8) being 62 lines Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an Abstract section. ** There are 6 instances of too long lines in the document, the longest one being 3 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the RFC 3978 Section 5.4 Copyright Line does not match the current year -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- Couldn't find a document date in the document -- date freshness check skipped. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'RFC 2119' is mentioned on line 192, but not defined == Unused Reference: 'KEY' is defined on line 354, but no explicit reference was found in the text == Unused Reference: 'ECRTP-MPLS-PROTO' is defined on line 374, but no explicit reference was found in the text ** Obsolete normative reference: RFC 3036 (ref. 'LDP') (Obsoleted by RFC 5036) -- Obsolete informational reference (is this intentional?): RFC 2547 (ref. 'MPLS-VPN') (Obsoleted by RFC 4364) Summary: 14 errors (**), 0 flaws (~~), 5 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Network Working Group Jerry Ash 2 Internet Draft Bur Goode 3 Category: Informational Jim Hand 4 AT&T 6 Raymond Zhang 7 Infonet Services Corporation 9 June, 2004 11 Requirements for Header Compression over MPLS 13 Status of this Memo: 15 By submitting this Internet-Draft, we certify that any applicable 16 patent or other IPR claims of which we are aware have been 17 disclosed, and any of which we become aware will be disclosed, in 18 accordance with RFC 3668 (BCP 79). 20 By submitting this Internet-Draft, we accept the provisions of 21 Section 3 of RFC 3667 (BCP 78). 23 Internet-Drafts are working documents of the Internet Engineering Task 24 Force (IETF), its areas, and its working groups. Note that other groups 25 may also distribute working documents as Internet-Drafts. 27 Internet-Drafts are draft documents valid for a maximum of six months 28 and may be updated, replaced, or obsoleted by other documents at any 29 time. It is inappropriate to use Internet-Drafts as reference material 30 or cite them other than as "work in progress". 32 The list of current Internet-Drafts can be accessed at 33 http://www.ietf.org/ietf/lid-abstracts.txt. 35 The list of Internet-Draft Shadow Directories can be accessed at 36 http://www.ietf.org/shadow.html. 38 This document is a submission of the IETF AVT WG. Comments should 39 be directed to the AVT WG mailing list, avt@ietf.org. 41 Abstract: 43 VoIP typically uses the encapsulation voice/RTP/UDP/IP. When MPLS 44 labels are added, this becomes voice/RTP/UDP/IP/MPLS-labels, where, for 45 example, the packet header is at least 48 bytes, while the voice payload 46 is often no more than 30 bytes. Header compression can significantly 47 reduce the overhead through various compression mechanisms, such as 48 enhanced compressed RTP (ECRTP) and robust header compression (ROHC). We 49 consider using MPLS to route compressed packets over an MPLS LSP without 50 compression/decompression cycles at each router. This approach can 51 increase the bandwidth efficiency as well as processing scalability of 52 the maximum number of simultaneous flows that use header compression at 53 each router. In the draft we give a problem statement, goals and 54 requirements, and an example scenario. 56 Table of Contents: 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 2 59 2. Problem Statement . . . . . . . . . . . . . . . . . . . . . . . 3 60 3. Goals & Requirements . . . . . . . . . . . . . . . . . . . . . . 4 61 4. Candidate Solution Methods & Needs . . . . . . . . . . . . . . . 5 62 5. Example Scenario . . . . . . . . . . . . . . . . . . . . . . . . 6 63 6. Security Considerations . . . . . . . . . . . . . . . . . . . . 7 64 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 7 65 8. Normative References . . . . . . . . . . . . . . . . . . . . . . 7 66 9. Informative References . . . . . . . . . . . . . . . . . . . . . 7 67 10. Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 8 68 11. Full Copyright Statement . . . . . . . . . . . . . . . . . . . 8 70 1. Introduction 72 Voice over IP (VoIP) typically uses the encapsulation voice/RTP/UDP/IP. 73 When MPLS labels [MPLS-ARCH] are added, this becomes 74 voice/RTP/UDP/IP/MPLS-labels. For an MPLS VPN (e.g., [MPLS-VPN], the 75 packet header is at least 48 bytes, while the voice payload is often no 76 more than 30 bytes, for example. The interest in header compression 77 (HC) is to exploit the possibility of significantly reducing the 78 overhead through various compression mechanisms, such as with enhanced 79 compressed RTP [ECRTP] or robust header compression [ROHC], and also to 80 increase scalability of HC. We consider using MPLS to route compressed 81 packets over an MPLS LSP (label switched path) without 82 compression/decompression cycles at each router. Such an HC over MPLS 83 capability can increase bandwidth efficiency as well as the processing 84 scalability of the maximum number of simultaneous flows which use HC at 85 each router. 87 To implement HC over MPLS, the ingress router/gateway would have to 88 apply the HC algorithm to the IP packet, the compressed packet routed on 89 an MPLS LSP using MPLS labels, and the compressed header would be 90 decompressed at the egress router/gateway where the HC session 91 terminates. Figure 1 illustrates an HC over MPLS session established on 92 an LSP that crosses several routers, from R1/HC --> R2 --> R3 --> R4/HD, 93 where R1/HC is the ingress router where HC is performed, and R4/HD is 94 the egress router where header decompression (HD) is done. HC of the 95 RTP/UDP/IP header is performed at R1/HC, and the compressed packets are 96 routed using MPLS labels from R1/HC to R2, to R3, and finally to R4/HD, 97 without further decompression/recompression cycles. The RTP/UDP/IP 98 header is decompressed at R4/HD and can be forwarded to other routers, 99 as needed. 100 _____ 101 | | 102 |R1/HC| Header Compression (HC) Performed 103 |_____| 104 | 105 | voice/compressed-header/MPLS-labels 106 V 107 _____ 108 | | 109 | R2 | 110 |_____| 111 | 112 | voice/compressed-header/MPLS-labels 113 V 114 _____ 115 | | 116 | R3 | 117 |_____| 118 | 119 | voice/compressed-header/MPLS-labels 120 V 121 _____ 122 | | 123 |R4/HD| Header Decompression (HD) Performed 124 |_____| 126 Figure 1. Example of Header Compression over MPLS over Routers R1-->R4 128 In the example scenario, HC therefore takes place between R1 and R4, and 129 the MPLS path transports voice/compressed-header/MPLS-labels instead of 130 voice/RTP/UDP/IP/MPLS-labels, typically saving 30 octets or more per 131 packet. The MPLS label stack and link-layer headers are not compressed. 132 A signaling method is needed to set up a correspondence between the 133 ingress and egress routers of the HC over MPLS session. 135 In Section 2 we give a problem statement, in Section 3 we give goals and 136 requirements, and in Section 4 we give an example scenario. 138 2. Problem Statement 140 As described in the introduction, HC over MPLS can significantly reduce 141 the header overhead through HC mechanisms. The need for HC may be 142 important on low-speed links where bandwidth is more scarce, but it 143 could also be important on backbone facilities, especially where costs 144 are high (e.g., some global cross-sections). VoIP typically will use 145 voice compression mechanisms (e.g., G.729) on low-speed and 146 international routes, in order to conserve bandwidth. With HC, 147 significantly more bandwidth could be saved. For example, carrying 148 uncompressed headers for the entire voice load of a large domestic 149 network with 300 million or more calls per day could consume on the 150 order of about 20-40 gigabits-per-second on the backbone network for 151 headers alone. This overhead could translate into considerable bandwidth 152 capacity. 154 The claim is often made that once fiber is in place, increasing the 155 bandwidth capacity is inexpensive, nearly 'free'. This may be true in 156 some cases, however, on some international cross-sections, especially, 157 facility/transport costs are very high and saving bandwidth on such 158 backbone links is very worthwhile. Decreasing the backbone bandwidth is 159 needed in some areas of the world where bandwidth is very expensive. It 160 is also important in almost all locations to decrease the bandwidth 161 consumption on low-speed links. So although bandwidth is getting 162 cheaper, the value of compression does not go away. It should be 163 further noted that IPv6 will increase the size of headers, and therefore 164 increase the importance of HC for RTP flows. 166 While hop-by-hop HC could be applied to decrease bandwidth requirements, 167 that implies a processing requirement for compression-decompression 168 cycles at every router hop, which does not scale well for large voice 169 traffic loads. The maximum number of cRTP flows is about 30-50 for a 170 typical customer premise router, depending upon its uplink speed and 171 processing power, while the need may exceed 300-500 for a high-end case. 172 Therefore, HC over MPLS seems to be a viable alternative to get the 173 compression benefits without introducing costly processing demands on 174 the intermediate nodes. By using HC over MPLS, routers merely forward 175 compressed packets without doing a decompression/recompression cycle, 176 thereby increasing the maximum number of simultaneous compressed flows 177 that routers can handle. 179 Therefore the proposal is to use existing HC techniques, together with 180 MPLS labels, to make the transport of the RTP/UDP/IP headers more 181 efficient over an MPLS network. However, at this time, there are no 182 standards for HC over MPLS, and vendors have not implemented such 183 techniques. 185 3. Goals & Requirements 187 Specification of Requirements 189 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 190 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 191 document are to be interpreted as described in [RFC 2119]. 193 The goals of HC over MPLS are as follows: 195 a. provide more efficient voice transport over MPLS networks, 196 b. increase the scalability of HC to a large number of flows, 197 c. not significantly increase packet delay, delay variation, or loss 198 probability, and 199 d. leverage existing work through use of standard protocols as much as 200 possible. 202 Therefore the requirements for HC over MPLS are as follows: 204 a. MUST use existing protocols (e.g., [ECRTP], [ROHC]) to compress 205 RTP/UDP/IP headers, in order to provide for efficient transport, 206 tolerance to packet loss, and resistance to loss of session context. 207 b. MUST allow HC over an MPLS LSP, and thereby avoid hop-by-hop 208 compression/decompression cycles [e.g., ECRTP-MPLS-PROTO]. 210 c. MUST minimize incremental performance degradation due to increased 211 delay, packet loss, and jitter. 212 d. MUST use standard protocols to signal context identification and 213 control information (e.g., [RSVP], [RSVP-TE], [LDP]). 214 e. Packet reordering MUST NOT cause incorrectly decompressed packets to 215 be forwarded from the decompressor. 217 It is necessary that the HC method be able to handle out-of-sequence 218 packets. MPLS [MPLS-ARCH] enables 4-byte labels to be appended to IP 219 packets to allow switching from the ingress label switched router (LSR) 220 to the egress LSP on an LSP through an MPLS network. However, MPLS does 221 not guarantee that packets will arrive in order at the egress LSR, since 222 a number of things could cause packets to be delivered out of sequence. 223 For example, a link failure could cause the LSP routing to change, due 224 perhaps to an MPLS fast reroute taking place, or to the interior gateway 225 protocol (IGP) and label distribution protocol (LDP) converging to 226 another route, among other possible reasons. Other causes could include 227 IGP reroutes due to 'loose hops' in the LSP, or BGP route changes 228 reflecting back into IGP reroutes. HC algorithms may be able to handle 229 reordering magnitudes on the order of about 10 packets, which may make 230 the time required for IGP reconvergence (typically on the order of 231 seconds) untenable for the HC algorithm. On the other hand, MPLS fast 232 reroute may be fast enough (on the order of 50 ms. or less) for the HC 233 algorithm to handle packet reordering. The issue of reordering needs to 234 be further considered in the development of the HC over MPLS solution. 236 Resynchronization and performance also needs to be considered, since HC 237 over MPLS can sometimes have multiple routers in the LSP. Tunneling a HC 238 session over an MPLS LSP with multiple routers in the path will increase 239 the round trip delay and the chance of packet loss, and HC contexts are 240 invalidated due to packet loss. The HC error recovery mechanism can 241 compound the problem when long round trip delays are involved. 243 4. Candidate Solution Methods & Needs 245 [cRTP] performs best with very low packet error rates on all hops of the 246 path. When the cRTP decompressor context state gets out of synch with 247 the compressor, it will drop packets associated with the context until 248 the two states are resynchronized. To resynchronize context state at the 249 two ends, the decompressor transmits the CONTEXT_STATE packet to the 250 compressor, and the compressor transmits a FULL_HEADER packet to the 251 decompressor. 253 [ECRTP] uses mechanisms that make cRTP more tolerant to packet loss, and 254 ECRTP thereby helps to minimize the use of feedback-based error recovery 255 (CONTEXT_STATE packets). ECRTP is therefore a candidate method to make 256 HC over MPLS more tolerant of packet loss and to guard against frequent 257 resynchronizations. ECRTP may need some implementation adaptations to 258 address the reordering requirement in Section 3 (requirement e), since 259 a default implementation will probably not meet the requirement. ECRTP 260 protocol extensions may be required to identify FULL_HEADER, CONTEXT_STATE, 261 and compressed packet types. [cRTP-ENCAP] specifies a separate link-layer 262 packet type defined for HC. Using a separate link-layer packet type avoids 263 the need to add extra bits to the compression header to identify the packet 264 type. However, this approach does not extend well to MPLS encapsulation 265 conventions [MPLS-ENCAP], in which a separate link-layer packet type 266 translates into a separate LSP for each packet type. In order to extend 267 ECRTP to HC over MPLS, each packet type defined in [ECRTP] would need to be 268 identified in an appended packet type field in the ECRTP header. 270 [ROHC] is also very tolerant of packet loss, and therefore is a 271 candidate method to guard against frequent resynchronizations. ROHC 272 also achieves a somewhat better level of compression as compared to 273 ECRTP. ROHC may need some implementation adaptations to address the 274 reordering requirement in Section 3 (requirement e), since a default 275 implementation will probably not meet the requirement. ROHC already has 276 the capability to identify the packet type in the compression header, so 277 no further extension is needed to identify packet type. 279 Extensions to MPLS signaling may be needed to identify the LSP from HC to 280 HD egress point, negotiate the HC algorithm used and protocol 281 parameters, and negotiate the session context IDs (SCIDs) space between 282 the ingress and egress routers on the MPLS LSP. For example, new 283 objects may need to be defined for [RSVP-TE] to signal the SCID spaces 284 between the ingress and egress routers, and the HC algorithm used to 285 determine the context; these HC packets then contain the SCID identified 286 by using the RSVP-TE objects. It is also desirable to signal HC over 287 MPLS tunnels with the label distribution protocol [LDP], since many 288 RFC2547 VPN [MPLS-VPN] implementations use LDP as the underlying LSP 289 signaling mechanism, and LDP is very scalable. However, extensions to 290 LDP may be needed to signal SCIDs between ingress and egress routers 291 on HC over MPLS LSPs. For example, 'targeted LDP sessions' might be 292 established for signaling SCIDs, or perhaps methods described in 293 [LDP-PWE3] and [GVPLS] to signal pseudo-wires and multipoint-to-point 294 LSPs might be extended to support signaling of SCIDs for HC over MPLS 295 LSPs. These MPLS signaling protocol extensions need coordination with 296 other working groups (e.g., MPLS). 298 5. Example Scenario 300 As illustrated in Figure 2, many VoIP flows are originated from customer 301 sites, which are served by routers R1, R2 and R3, and terminated at 302 several large customer call centers, which are served by R5, R6 and R7. 303 R4 is a service-provider router, and all VoIP flows traverse R4. It is 304 essential that the R4-R5, R4-R6, and R4-R7 low-speed links all use HC to 305 allow a maximum number of simultaneous VoIP flows. To allow processing 306 at R4 to handle the volume of simultaneous VoIP flows, it is desired to 307 use HC over MPLS for these flows. With HC over MPLS, R4 does not need 308 to do HC/HD for the flows to the call centers, enabling more scalability 309 of the number of simultaneous VoIP flows with HC at R4. 311 voice/C-HDR/MPLS-labels ______ voice/C-HDR/MPLS-labels 312 R1/HC---------------------->| |-----------------------> R5/HD 313 | | 314 voice/C-HDR/MPLS-labels| |voice/C-HDR/MPLS-labels 315 R2/HC---------------------->| R4 |-----------------------> R6/HD 316 | | 317 voice/C-HDR/MPLS-labels| |voice/C-HDR/MPLS-labels 318 R3/HC---------------------->|______|-----------------------> R7/HD 320 [Note: HC = header compression; C-HDR = compressed header; HD = 321 header decompression] 323 Figure 2. Example Scenario for Application of HC over MPLS 325 6. Security Considerations 327 The high processing load of HC makes HC a target for denial-of-service 328 attacks. For example, an attacker could send a high bandwidth data 329 stream through a network, with the headers in the data stream marked 330 appropriately to cause HC to be applied. This would use large amounts 331 of processing resources on the routers performing compression and 332 decompression, and these processing resources might then be unavailable 333 for other important functions on the router. This threat is not a new 334 threat for HC, but is addressed and mitigated by HC over MPLS. That is, 335 by reducing the need for performing compression and decompression 336 cycles, as proposed in this draft, the risk of this type of 337 denial-of-service attack is reduced. 339 7. IANA Considerations 341 No IANA actions are required. 343 8. Normative References 345 [cRTP] Casner, S., Jacobsen, V., "Compressing IP/UDP/RTP Headers for 346 Low-Speed Serial Links", RFC 2508, February 1999. 348 [cRTP-ENCAP] Engan, M., Casner, S., Bormann, C., "IP Header Compression 349 over PPP", RFC 2509, February 1999. 351 [ECRTP] Koren, T., et. al., "Compressing IP/UDP/RTP Headers on Links 352 with High Delay, Packet Loss, and Reordering," RFC 3545, July 2003. 354 [KEY] Bradner, S., "Key words for use in RFCs to Indicate Requirement 355 Levels", RFC 2119, March 1997. 357 [LDP] Andersson, L., et. al., "LDP Specification", RFC 3036, January 358 2001. 360 [MPLS-ARCH] Rosen, E., et. al., "Multiprotocol Label Switching 361 Architecture," RFC 3031, January 2001. 363 [ROHC] Bormann, C., et. al., "Robust Header Compression (ROHC)," RFC 364 3091, July 2001. 366 [RSVP] Braden, R. et al., "Resource ReSerVation Protocol (RSVP) -- 367 Version 1, Functional Specification", RFC 2205, September 1997. 369 [RSVP-TE] Awduche, D., et. al., "RSVP-TE: Extensions to RSVP for LSP 370 Tunnels", RFC 3209, December 2001. 372 9. Informative References 374 [ECRTP-MPLS-PROTO] Ash, G., Goode, B., Hand, J., "Protocol Extensions 375 for Header Compression over MPLS", work in progress. 377 [GVPLS] Radoaca, V., et. al., "GVPLS/LPE - Generalized VPLS Solution 378 based on LPE Framework," work in progress. 380 [LDP-PWE3] Martini, L., et. al., "Pseudowire Setup and Maintenance using 381 LDP", work in progress. 383 [MPLS-ENCAP] Rosen, E., et. al., "MPLS Label Stack Encoding", RFC 3032, 384 January 2001. 386 [MPLS-VPN] Rosen, E., Rekhter, Y., "BGP/MPLS VPNs", RFC 2547, March 387 1999. 389 10. Authors' Addresses 391 Jerry Ash 392 AT&T 393 Room MT D5-2A01 394 200 Laurel Avenue 395 Middletown, NJ 07748, USA 396 Phone: +1 732-420-4578 397 Email: gash@att.com 399 Bur Goode 400 AT&T 401 Phone: + 1 203-341-8705 402 E-mail: bgoode@att.com 404 Jim Hand 405 Consultant 406 E-mail: hand17@earthlink.net 408 Raymond Zhang 409 Infonet Services Corporation 410 2160 E. Grand Ave. El Segundo, CA 90025 USA 411 Email: zhangr@infonet.com 413 11. Full Copyright Statement 415 Copyright (C) The Internet Society (2004). This document is subject to 416 the rights, licenses and restrictions contained in BCP 78 and except as 417 set forth therein, the authors retain all their rights. 419 This document and the information contained herein are provided on an 420 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR 421 IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET 422 ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, 423 INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE 424 INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 425 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.