idnits 2.17.1 draft-jiang-asymmetric-ipv6-04.txt: -(411): Line appears to be too long, but this could be caused by non-ascii characters in UTF-8 encoding Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == There is 1 instance of lines with non-ascii characters in the document. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (15 November 2020) is 1258 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 6145 (Obsoleted by RFC 7915) Summary: 2 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group S. Jiang 3 Internet-Draft Huawei Technologies Co., Ltd 4 Intended status: Informational G. Li 5 Expires: 19 May 2021 Huawei Technologies 6 B. E. Carpenter 7 Univ. of Auckland 8 15 November 2020 10 Asymmetric IPv6 for Resource-constrained IoT Networks 11 draft-jiang-asymmetric-ipv6-04 13 Abstract 15 This document describes a new approach to IPv6 header compression for 16 use in scenarios where minimizing packet size is crucial but routing 17 performance must be maximised. 19 Status of This Memo 21 This Internet-Draft is submitted in full conformance with the 22 provisions of BCP 78 and BCP 79. 24 Internet-Drafts are working documents of the Internet Engineering 25 Task Force (IETF). Note that other groups may also distribute 26 working documents as Internet-Drafts. The list of current Internet- 27 Drafts is at https://datatracker.ietf.org/drafts/current/. 29 Internet-Drafts are draft documents valid for a maximum of six months 30 and may be updated, replaced, or obsoleted by other documents at any 31 time. It is inappropriate to use Internet-Drafts as reference 32 material or to cite them other than as "work in progress." 34 This Internet-Draft will expire on 19 May 2021. 36 Copyright Notice 38 Copyright (c) 2020 IETF Trust and the persons identified as the 39 document authors. All rights reserved. 41 This document is subject to BCP 78 and the IETF Trust's Legal 42 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 43 license-info) in effect on the date of publication of this document. 44 Please review these documents carefully, as they describe your rights 45 and restrictions with respect to this document. Code Components 46 extracted from this document must include Simplified BSD License text 47 as described in Section 4.e of the Trust Legal Provisions and are 48 provided without warranty as described in the Simplified BSD License. 50 Table of Contents 52 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 53 2. Proposed Solution . . . . . . . . . . . . . . . . . . . . . . 3 54 3. Address Transformation at the Gateway . . . . . . . . . . . . 6 55 4. Routing without Decompression . . . . . . . . . . . . . . . . 6 56 5. Address Configuration . . . . . . . . . . . . . . . . . . . . 7 57 6. Compatibility with Existing Protocols . . . . . . . . . . . . 7 58 7. Relationship to Static Context Header Compression . . . . . . 7 59 8. Security Considerations . . . . . . . . . . . . . . . . . . . 8 60 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 61 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 8 62 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 8 63 Appendix A. Change log [RFC Editor: Please remove] . . . . . . . 10 64 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 10 66 1. Introduction 68 The large address space of IPv6 is essential for the massive 69 expansion of the network edge that will be caused by "Internet of 70 Things" (IoT) technology over low-power or 5G links. However, the 71 size of a raw IPv6 packet header causes difficulty due to the small 72 maximum transmission units (MTU) allowed by typical low-power, low- 73 cost link layers. For 5G, the importance of header overhead in small 74 packets is discussed in [NGMN-5G]. Thus header compression, 75 including address compression, is an important issue. This decreases 76 the size of raw packets, but compressed IP addresses are not 77 routeable except by decompressing them completely in every forwarding 78 node. There are two issues here. The first is the extra computation 79 resource needed for compressing or decompressing in constrained IoT 80 nodes. The second is that full-length IPv6 routing will consume more 81 memory to store routing tables and packet queues (assuming that 82 routing is not bypassed by tunnelling). Such resource consumption is 83 very undesirable in constrained nodes with limited storage, CPU 84 power, and battery capacity. 86 To mitigate these issues, here we propose a solution enabling the 87 shortening of IPv6 addresses inside packets, and the routing of 88 packets according to short addresses, without needing the overhead of 89 a decompression step prior to route lookup. Considering that the 90 scale and size of edge networks may vary widely, different lengths of 91 short address can be used in different domains. 93 As an illustrative example, consider an edge network which is known 94 to never require more than a few hundred nodes, which in most cases 95 will communicate either with each other, or with application layer 96 gateways to the rest of the Internet. Rather than needing 128-bit 97 addresses, such a network could very well operate with 16-bit 98 addresses. Also, it could very likely operate without needing 99 enhancements such as differentiated services, ECN or flow labels. If 100 only IPv6 is supported, the version number field is pointless. There 101 is no reason for IPv6 packets within such a network to contain 102 40-byte headers as specified in [RFC8200]. Therefore, the useful 103 information could be carried in 8 bytes (see Figure 1). Furthermore, 104 routers within the edge network can route packets directly on 16-bit 105 addresses, reducing RIB and FIB sizes and the lookup time. 107 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 108 | Payload Length | Next Header | Hop Limit | 109 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 110 | Source Address | Destination Address | 111 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 113 Figure 1 115 This work is distinct from previous work on address compression 116 [RFC6282] [RFC7400]. Although those solutions tackle the problem of 117 small MTU size, they do not address the problem of decompression 118 overhead. 120 This work is also distinct from the work on static context header 121 compression [RFC8724], as discussed in more detail below. 123 Finally, this work is distinct from the 6LoWPAN Routing Header 124 [RFC8138], which can support truncated addresses in a different way. 126 2. Proposed Solution 128 The use of IPv6 naturally implies 128-bit addresses for both source 129 and destination. However, this address size is huge by the standards 130 of IoT edge networks. We propose the use of a context parameter to 131 indicate the effective length of the IP address for every node in a 132 local domain. If the effective length is N bits, then all addresses 133 in the domain are assumed to be preceded by a common prefix of 128-N 134 bits, when a full size IPv6 address is needed. Any node in the 135 domain that needs the full address, such as a gateway node to the 136 Internet, can therefore easily synthesize it. If a client 137 communicates with a server that is in the local domain, short 138 addresses will be used end-to-end. 140 The address length parameter may be needed by every node in the 141 domain. It can be spread by various techniques: 143 * Configure the address length in every node. 145 * Obtain the address length from a gateway (next hop router) node. 147 * Negotiate the address length between neighbors. 149 The solution operates by shortening IP address fields to save 150 overhead. To enhance this, we propose a new field named Flexible 151 Header Encoding (FHE). It consists of 8 bits, each indicating 152 whether the corresponding IPv6 header field [RFC8200] exists. 154 * Bit 0 indicates the Modified Version field 156 * Bit 1 indicates the Traffic Class field 158 * Bit 2 indicates the Flow Label field. 160 * Bit 3 indicates the Payload Length field. 162 * Bit 4 indicates the Next Header field. (Zero implies "No Next 163 Header", value 59) 165 * Bit 5 indicates the Hop Limit field. 167 * Bit 6 indicates the Source Address field. 169 * Bit 7 indicates the Destination Address field. 171 The "Version" field is a special case. In the context of FHE, all 172 packets are presumed to be IPv6 so the normal version field has no 173 purpose. The Modified Version field, if present, has the following 174 encoded meanings: 176 * 0b0000: The source address (if exist) has pre-determined length 177 inside the domain and the destination address (if exist) uses 178 standard 128-bit IPv6 address. (Outward traffic) 180 * 0b0001: The source address (if exist) uses standard 128-bit IPv6 181 address and the destination address (if exist) has pre-determined 182 length inside the domain. (Inward traffic) 184 * 0b0010: The source address and destination address have the same 185 length inside the domain. The address length will be pre- 186 determined. 188 * 0b0110: Reserved for IPv6 compatible case. 190 * 0b0100: Reserved for IPv4 compatible case. 192 * 0b0011~0b1111(except 0b0110, 0b0100): Reserved. 194 All fields, including the Modified Version field, follow the FHE in 195 the same order as in [RFC8200], with no padding. There are no 196 alignment requirements, but when a packet is decompressed to a normal 197 IPv6 format, padding options as defined in RFC8200 must be inserted. 199 Compared to the illustrative example in Figure 1, the actual packet 200 size would therefore be 10 bytes, a considerable improvement on the 201 standard 40 bytes. 203 One implication of the above is that the source and destination 204 addresses may be elided completely if they are implicit. Sourceless 205 packets were originally suggested in [Crowcroft]. 207 Figure 2 illustrates an example of the FHE format. In this example 208 the traffic class, flow label and source address are elided, and the 209 destination address is truncated to 16 bits. The modified version 210 field could be 0b0001 or 0b0010. 212 FHE octet 213 Modified +-+-+-+-+-+-+-+-+ 214 Version |1 0 0 1 1 1 0 1| 215 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 216 |0 0 0 1| Payload Length | Next Header | Hop | 217 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 218 | Limit | Truncated Destination Address | | 219 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 220 | | 221 + Transport payload | 222 | | 223 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+......... 225 Figure 2 227 Note that Asymmetric IPv6 does not contain any special handling for 228 IPv6 fragmentation, which will operate exactly as described in 229 [RFC8200], with Asymmetric IPv6 applied to each fragment packet. 230 However, we assume that in IoT deployment scenarios, packets whose 231 length exceeds the IPv6 minimum link MTU before applying Asymmetric 232 IPv6 will be rare. If the underlying link layer cannot carry 233 complete packets even after applying Asymmetric IPv6 compression, an 234 adaptation layer will be necessary exactly as for normal IPv6. 236 3. Address Transformation at the Gateway 238 Truncated intra-domain addresses will be used to identify nodes 239 inside the domain. When a packet is sent from an IoT node to an 240 external IPv6 host , the node's intra-domain address, which is unique 241 in the domain, will be carried in the source address field. When the 242 packet is forwarded outside the domain by a gateway, the intra-domain 243 address will be transformed to a complete IPv6 address. To achieve 244 this, the gateway should will maintain a globally routeable prefix 245 for all the nodes in the domain. When a packet with an intra-domain 246 source address is received, the gateway extracts this address and 247 concatenates it to the prefix to form a standard, globally unique 248 IPv6 address. Vice versa, when IPv6 packets are received from the 249 Internet, the prefix will be removed to recover the intra-domain 250 short address. 252 There are two options for handling the addresses of external hosts 253 within the domain. One is to use their full IPv6 addresses via 254 Modified Version codes 0b0000 and 0b0001. The other is effectively a 255 specialized form of Network Address Translation. Here, the gateway 256 will maintain a dynamic mapping table between synthetic intra-domain 257 addresses and IPv6 addresses. As packets are received, the gateway 258 performs the appropriate mapping. The transformation must be 259 checksum-neutral for the transport layer, so the methods designed for 260 NAT46 should be adapted [RFC6145]. 262 It is an engineering choice whether this method is preferable to 263 carrying full 128-bit addresses on the IOT side. Which type of 264 resource is more expensive should be seriously considered to choose 265 the appropriate ways, e.g. computing, memory, or transmiting in 266 various resource-constrained IoT networks. 268 4. Routing without Decompression 270 Routing mechanisms may readily be adapted to truncated address sizes. 271 If there is routing with an HFE domain, we assume that the truncated 272 address size will be split into a prefix and an interface identifier, 273 but this will not be at the traditional /64 boundary. If routing 274 between different length addresses is required, a suitably modified 275 Forwarding Information Base (FIB) structure is needed, as for any 276 variable length addressing scheme. A truncated address needs to be 277 virtually expanded to 128 bits at the router's inbound interface, 278 although this may not be the physical implementation. 280 A possible routing choice for IOT edge networks is RPL [RFC6550], 281 although a more complete survey can be found in [Talwar]. 283 5. Address Configuration 285 The simplest approach to address configuration is simply to run 286 normal IPv6 procedures (SLAAC or DHCPv6), on the argument that this 287 is a rare process and the overhead does not matter. If the truncated 288 address size is less than 64 bits, it will be necessary to use 289 shorter interface identifiers than normal, but this is not a major 290 change. Once a node has acquired an IPv6 address and has learned the 291 local address length parameter as outlined in Section 2, it can 292 continue in FHE mode. 294 6. Compatibility with Existing Protocols 296 Although HFE nodes can only talk directly to each other, they are 297 essentially a special form of IPv6 node and they can communicate with 298 the whole IPv6 Internet via gateways. The complexity is not greater 299 than 6LoWPAN. If appropriate, the 6LoWPAN adaptation layer [RFC4944] 300 could be used, with a specific dispatch type. 302 7. Relationship to Static Context Header Compression 304 Static Context Header Compression (SCHC) [RFC8724] is a powerful 305 mechanism for reducing IPv6 packet size in an IoT application 306 environment. In particular it includes a profile for UDP over IPv6, 307 and a somewhat modified version of this profile could achieve much of 308 what Asymmetric IPv6 proposes. In addition, SCHC provides support 309 for fragmentation in the case of very small link MTUs. However, SCHC 310 is by design static, and once a context is established the fields to 311 be compressed do not change. Asymmetric IPv6 transmits the FHE and 312 Modified Version bytes with every packet, so it provides dynamic 313 choice as to which header elements are compressed or elided. 315 In a context where the desirable compression is fixed, e.g. every 316 address is the same length, the flow label is never used, etc., SCHC 317 can used to the same effect as Asymmetric IPv6. However, if the 318 behavior needs to be dynamic, the signaling power of the FHE and 319 Modified Version bytes in Asymmetric IPv6 is needed. 321 Further study is needed whether the advantages of the two mechanisms 322 can be combined. 324 8. Security Considerations 326 HFE is essentially only a non-cryptographic compression technique so 327 it neither adds to nor reduces the intrinsic security of an IPv6 328 packet. The address length parameter is not a secret, since all 329 nodes in the domain must know it. The mechanism for distributing 330 this parameter must be no less secure than any other configuration 331 mechanism in us. 333 Address-based privacy issues must be considered in deciding on the 334 address length. If the number of bits available for the interface 335 identifier is significantly less than the 64 currently in use, 336 address traceability and guessability will be affected. However, if 337 the traffic with short addresses is confined to within the edge 338 network, the privacy issue will be minimized. [RFC7721] and 339 [RFC7217] should be consulted prior to deciding the address length. 341 9. IANA Considerations 343 This document makes no request of the IANA. 345 NOTE IN DRAFT: If the solution of a 6LoWPAN dispatch type is adopted, 346 a suitable assignment request will be added. 348 10. Acknowledgements 350 Useful comments were received from Uma Chunduri, Cheng Li, Pascal 351 Thubert, Laurent Toutain and others. 353 11. References 355 [Crowcroft] 356 Crowcroft, J. and M. Bagnulo, "SNA: Sourceless Network 357 Architecture", University of Cambridge Computer Laboratory 358 Technical Report UCAM-CL-TR-849, 2014. 360 [NGMN-5G] Thibault, I., "5G Extreme Requirements: Operators' views 361 on fundamental trade-offs", NGMN Alliance , 2017. 363 [RFC4944] Montenegro, G., Kushalnagar, N., Hui, J., and D. Culler, 364 "Transmission of IPv6 Packets over IEEE 802.15.4 365 Networks", RFC 4944, DOI 10.17487/RFC4944, September 2007, 366 . 368 [RFC6145] Li, X., Bao, C., and F. Baker, "IP/ICMP Translation 369 Algorithm", RFC 6145, DOI 10.17487/RFC6145, April 2011, 370 . 372 [RFC6282] Hui, J., Ed. and P. Thubert, "Compression Format for IPv6 373 Datagrams over IEEE 802.15.4-Based Networks", RFC 6282, 374 DOI 10.17487/RFC6282, September 2011, 375 . 377 [RFC6550] Winter, T., Ed., Thubert, P., Ed., Brandt, A., Hui, J., 378 Kelsey, R., Levis, P., Pister, K., Struik, R., Vasseur, 379 JP., and R. Alexander, "RPL: IPv6 Routing Protocol for 380 Low-Power and Lossy Networks", RFC 6550, 381 DOI 10.17487/RFC6550, March 2012, 382 . 384 [RFC7217] Gont, F., "A Method for Generating Semantically Opaque 385 Interface Identifiers with IPv6 Stateless Address 386 Autoconfiguration (SLAAC)", RFC 7217, 387 DOI 10.17487/RFC7217, April 2014, 388 . 390 [RFC7400] Bormann, C., "6LoWPAN-GHC: Generic Header Compression for 391 IPv6 over Low-Power Wireless Personal Area Networks 392 (6LoWPANs)", RFC 7400, DOI 10.17487/RFC7400, November 393 2014, . 395 [RFC7721] Cooper, A., Gont, F., and D. Thaler, "Security and Privacy 396 Considerations for IPv6 Address Generation Mechanisms", 397 RFC 7721, DOI 10.17487/RFC7721, March 2016, 398 . 400 [RFC8138] Thubert, P., Ed., Bormann, C., Toutain, L., and R. Cragie, 401 "IPv6 over Low-Power Wireless Personal Area Network 402 (6LoWPAN) Routing Header", RFC 8138, DOI 10.17487/RFC8138, 403 April 2017, . 405 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 406 (IPv6) Specification", STD 86, RFC 8200, 407 DOI 10.17487/RFC8200, July 2017, 408 . 410 [RFC8724] Minaburo, A., Toutain, L., Gomez, C., Barthel, D., and JC. 411 Zúñiga, "SCHC: Generic Framework for Static Context Header 412 Compression and Fragmentation", RFC 8724, 413 DOI 10.17487/RFC8724, April 2020, 414 . 416 [Talwar] Talwar, M., "Routing Techniques and Protocols for Internet 417 of Things: a Survey", Indian J.Sci.Res. 12(1):417-423, 418 2015. 420 Appendix A. Change log [RFC Editor: Please remove] 422 * draft-jiang-asymmetric-ipv6-00, 2019-06-03: 424 - Initial version 426 * draft-jiang-asymmetric-ipv6-01, 2019-06-21: 428 - Fixed reference error 430 * draft-jiang-asymmetric-ipv6-02, 2019-10-29: 432 - Added illustrative example 434 - Discussed fragmentation 436 - Discussed relationship to SCHC 438 - Fixed bit pattern errors 440 * draft-jiang-asymmetric-ipv6-03, 2020-05-15: 442 - Minor technical and editorial fixes 444 - Converted to xml2rfc v3 446 * draft-jiang-asymmetric-ipv6-04, 2020-11-15: 448 - Explicitly limit the scope to resource-constrained domain 450 - Add engineering choice considerations accordingly 452 Authors' Addresses 454 Sheng Jiang 455 Huawei Technologies Co., Ltd 456 Q14, Huawei Campus, No.156 Beiqing Road 457 Hai-Dian District, Beijing, 100095 458 P.R. China 460 Email: jiangsheng@huawei.com 462 Guangpeng Li 463 Huawei Technologies 464 Q14, Huawei Campus 465 No.156 Beiqing Road 466 Hai-Dian District, Beijing 467 100095 468 P.R. China 470 Email: liguangpeng@huawei.com 472 Brian Carpenter 473 The University of Auckland 474 School of Computer Science 475 University of Auckland 476 PB 92019 477 Auckland 1142 478 New Zealand 480 Email: brian.e.carpenter@gmail.com