idnits 2.17.1 draft-wood-dtnrg-http-dtn-delivery-06.txt: 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: ---------------------------------------------------------------------------- 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 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.) -- The document date (November 14, 2010) is 4909 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 2616 (Obsoleted by RFC 7230, RFC 7231, RFC 7232, RFC 7233, RFC 7234, RFC 7235) == Outdated reference: A later version (-09) exists of draft-irtf-dtnrg-bundle-checksum-08 == Outdated reference: A later version (-22) exists of draft-wood-tsvwg-saratoga-06 -- Obsolete informational reference (is this intentional?): RFC 2818 (Obsoleted by RFC 9110) Summary: 1 error (**), 0 flaws (~~), 3 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group L. Wood 3 Internet-Draft University of Surrey 4 Intended status: Experimental P. Holliday 5 Expires: May 18, 2011 Cisco Systems 6 November 14, 2010 8 Using HTTP for delivery in Delay/Disruption-Tolerant Networks 9 draft-wood-dtnrg-http-dtn-delivery-06 11 Abstract 13 This document describes how to use the Hypertext Transfer Protocol, 14 HTTP, for communication across delay- and disruption-tolerant 15 networks, by making every transit node in the network HTTP-capable, 16 and doing peer HTTP transfers between nodes to move data hop-by-hop 17 or subnet-by-subnet towards its final destination. HTTP is well- 18 known and straightforward to implement in these networks. 20 Status of this Memo 22 This Internet-Draft is submitted to IETF in full conformance with the 23 provisions of BCP 78 and BCP 79. 25 Internet-Drafts are working documents of the Internet Engineering 26 Task Force (IETF). Note that other groups may also distribute 27 working documents as Internet-Drafts. The list of current Internet- 28 Drafts is at http://datatracker.ietf.org/drafts/current/. 30 Internet-Drafts are draft documents valid for a maximum of six months 31 and may be updated, replaced, or obsoleted by other documents at any 32 time. It is inappropriate to use Internet-Drafts as reference 33 material or to cite them other than as "work in progress." 35 This Internet-Draft will expire on May 18, 2011. 37 Copyright Notice 39 Copyright (c) 2010 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents 44 (http://trustee.ietf.org/license-info) in effect on the date of 45 publication of this document. Please review these documents 46 carefully, as they describe your rights and restrictions with respect 47 to this document. Code Components extracted from this document must 48 include Simplified BSD License text as described in Section 4.e of 49 the Trust Legal Provisions and are provided without warranty as 50 described in the Simplified BSD License. 52 Table of Contents 54 1. Background and Introduction . . . . . . . . . . . . . . . . . 3 55 2. Adapting the HTTP delivery mechanism for DTNs . . . . . . . . 5 56 3. Other useful proposed additional HTTP headers . . . . . . . . 7 57 4. Other suggestions on using MIME in DTN networks . . . . . . . 8 58 5. Security Considerations . . . . . . . . . . . . . . . . . . . 9 59 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 60 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 9 61 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 10 62 8.1. Normative References . . . . . . . . . . . . . . . . . . . 10 63 8.2. Informative References . . . . . . . . . . . . . . . . . . 10 64 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 12 66 1. Background and Introduction 68 Delay- and Disruption-Tolerant Networks (DTNs) are networks where 69 conditions are such that links between nodes are not always 70 permanent, may be of very long delay or exist only during very short 71 contact periods where the link is up, and may change over time 72 [RFC4838]. Some DTNs can be thought of as sparse ad-hoc networks, 73 with nodes communicating intermittently only when they come into 74 contact. Store-and-forward delivery of data is a useful way of 75 communicating across these networks. 77 A specialised store-and-forward protocol for DTN delivery has been 78 proposed in the IRTF DTN research group (DTNRG) - the Bundle Protocol 79 [RFC5050]. Criticisms of the Bundle Protocol's lack of reliability 80 and its complexity have been made [I-D.irtf-dtnrg-bundle-checksum]. 81 The Bundle Protocol is itself intended to be a routable data format, 82 but the supporting architectures for node and application naming/ 83 addressing, automated routing, security, QoS, and resource discovery 84 have not yet been agreed upon or in some cases even significantly 85 worked on. These things already exist for the Internet Protocol, and 86 can in many cases be easily leveraged for DTN networks [Wood09a]. 88 This document outlines how the well-known Hypertext Transfer Protocol 89 (HTTP) [RFC2616] can be used for store-and-forward communication 90 across DTNs. HTTP is not used end-to-end as it is on the web. 91 Instead, applications running on each node in the network communicate 92 with their neighbours using dedicated hop-by-hop or subnet-by-subnet 93 HTTP transfers to effect local data delivery. Additional HTTP header 94 information adds context for onward forwarding and delivery to 95 destination endpoints, and provides the reliability and support for 96 error-detection currently missing from the alternative Bundle 97 Protocol. 99 It must be stressed that this proposed use is distinct from proxy 100 caching methods prevalent in the traditional web. Caching commands 101 are not used; end-to-end HTTP requests are not intercepted by 102 intermediate caches that attempt to fulfil them in the traditional 103 web caching sense. 105 Although HTTP-DTN use as as a hop-by-hop message carrier between 106 caches implementing some form of routing protocol between them, the 107 distinction between client, server and proxy is replaced by peer 108 intermediate caches using HTTP to communicate in separate sessions 109 that together combine over time to make the full path between 110 original source and final destination for the data. 112 HTTP is a session layer, running over a transport layer providing 113 reliable delivery of the HTTP stream between hops. This transport 114 layer is commonly (and almost universally) TCP in the terrestrial 115 Internet, although alternative transport layers, such as SCTP, can 116 also be used under HTTP [I-D.natarajan-http-over-sctp]. For long- 117 delay networks, or for network conditions where TCP or an equivalent 118 is not suitable, an alternative transport layer such as Saratoga 119 [I-D.wood-tsvwg-saratoga] can be used under HTTP instead in hop-by- 120 hop communications between nodes. HTTP requires only reliable 121 streaming that can be used to provide ordered delivery; how that 122 reliable streaming is provided is up to the local transport layer in 123 the local subnet, and multiple different transport layers can be used 124 across the multiple hops between nodes to transfer data from source 125 to final destination. 127 Steve Deering has often described IP as 'the waist in the hourglass' 128 [Deering98] - what is above and touching on IP can be changed, what 129 is below and touching on IP can be changed, but provided the new 130 elements continue to interface to and work with IP, the hourglass 131 remains complete and the network stack remains functional. Here, 132 HTTP is the waist in this particular hourglass; applications can use 133 HTTP to communicate, provided HTTP runs over a reliable transport 134 stream. The applications can vary. The transport stream can be 135 changed; HTTP does not have to run over TCP/IP, but could even be 136 made to run directly over e.g. HDLC or a CCSDS reliable bitstream. 137 Given the prevalence of IP in many networks, it is likely that two 138 waists exist; IP and HTTP are likely choices, but the transport 139 protocol and physical enviroment will vary more. An expansion of 140 this argument is given in [Wood09b] 142 Separation of HTTP from the underlying transport layer to make HTTP a 143 layer in its own right is increasingly likely to happen; this is 144 analogous to the use of different "convergence layers" under the 145 Bundle Protocol. Being able to set what transport layer to use 146 depending on conditions is useful, and one simple configuration 147 approach to this, able to support HTTP-DTN, was outlined in 148 [I-D.wood-tae-specifying-uri-transports]. 150 HTTP use here relies on the three P's - Persistence, Pipelining and 151 the PUT directive. These are all present in the HTTP/1.1 152 specification. 154 This document contains an overview of how HTTP can be simply adapted 155 to the DTN environment by the use of HTTP/1.1 with persistence and 156 pipelining, the PUT and GET directives, and some trivial extra HTTP 157 headers needed to indicate e.g. a destination in the DTN network. 159 The remainder of this specification uses 'file' as a shorthand for 160 'binary object', which may be an HTTP 'object', file with an 161 associated MIMEtype, or other type of contiguous binary data. 163 A significant benefit to use of HTTP is that the well-known MIMEtype 164 mechanism, integral to HTTP, provides hints on what received files 165 are, and what applications should do with them [RFC2045]. The Bundle 166 Protocol does not support MIMEtypes, or any similar mechanism. HTTP/ 167 1.1's use of MIME is specified in [RFC2616] rather than in the 168 separate MIME documents. 170 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 171 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 172 document are to be interpreted as described in RFC 2119. [RFC2119] 174 2. Adapting the HTTP delivery mechanism for DTNs 176 Here, HTTP is used as a peer-to-peer protocol in the sense that 177 multiple files may be transferred in both directions simultaneously 178 between two communicating nodes using HTTP for DTN use. There is not 179 intended to be a strict client/user-agent to server relationship as 180 there is in the web. Instead, sending data across a path of six 181 nodes, four nodes between source and destination, will require a 182 minimum of five separate per-hop HTTP transactions between each pair 183 of nodes to move the data onwards to the next node. This breaks the 184 traditional end-to-end control loop and transfer into separate 185 control loops and transfers suitable for the DTN environment. 187 When two nodes come into contact across a local hop or a subnet, a 188 request for files to be copied, stored, and carried onwards can be 189 made by the receiving node issuing an HTTP GET request. 190 Alternatively, the sending node can simply issue a series of HTTP PUT 191 requests once a connection is established, if it believes that 192 putting the data to the receiving node moves it closer to its 193 eventual destination. The receiving node can always reject transfers 194 with error codes. 196 HTTP-DTN is a superset of HTTP/1.1. HTTP/1.1 pipelining and 197 persistence permits multiple PUTs to be made in sequence. Support 198 for these in implementations is crucial to the mechanisms outlined 199 here. (Note that [I-D.natarajan-http-over-sctp] also takes advantage 200 of HTTP pipelining and persistence.) 202 The key to enabling HTTP use for DTN networking is an added Content- 203 Destination: header, which specifies the final destination of the 204 file, and can be used by routing in the HTTP-using applications to 205 decide over which available links the file should be sent. Content-* 206 headers are special, in that they may not be ignored (section 9.6 of 207 [RFC2616]). Recipients not understanding Content-Destination: will 208 generate a "501 (Not Implemented)" error code. This separates HTTP 209 use in DTNs described here from normal end-to-end HTTP web use. HTTP 210 DTN nodes MUST support Content-Destination:. Files that are PUT are 211 cached and then relayed onwards by intermediate peers to the Content- 212 Destination:. GET requests for files can be forwarded by 213 intermediate peers to the Content-Destination:. 215 The information provided in Content-Destination: identifying the 216 destination may be an IP address, DNS name, Bundle Endpoint 217 Identifier (EID) or other text-string identifier useful to the local 218 DTN routing mechanisms being used. 220 Similarly, a Content-Source: header provides a textual identification 221 of the original source of the data. HTTP-DTN nodes MUST support 222 Content-Source:. 224 For DTN use, DTN HTTP nodes MUST also implement and use Content- 225 Length: and Content-Range: headers. These permit partial delivery of 226 files and resends of missing pieces of files. The Content-MD5: 227 header must be supported. This provides a simple end-to-end 228 reliability check. The Content-MD5: header is intended to be 229 generated by the source node first sending the data, and is not 230 recomputed at other nodes. 232 DTN HTTP nodes MUST implement the Host: header, in line with current 233 HTTP specifications. This header field MAY be left blank to request 234 available files from the peer node, rather than identifying a desired 235 file from a distant source by hostname matching the advertised 236 Content-Source: header. A sender placing a new file into the DTN 237 network for onward transmission MUST have the Content-Source: field 238 of the data being sent match its Host: field. 240 Hop-by-hop HTTP headers MAY be implemented between peer nodes talking 241 directly. The headers described in section 13.5.1 of [RFC2616] are 242 available. New hop-by-hop headers MUST use the Connection: header 243 approach described in section 14.10 of [RFC2616]. 245 DTN HTTP nodes may optionally GET from and PUT to link-local IP 246 multicast addresses when used over IP subnets. This permits 247 efficient sharing of files on shared LANs, with recipients requesting 248 resends via Content-Range: and checking assembly of file pieces using 249 the Content-MD5: header. A GET to multicast can request a specific 250 file from any available node that has it. The response to a 251 multicast GET SHOULD be unicast, but a multicast HEAD MAY also be 252 sent to inform other nodes that the sender has the file of interest. 253 If other nodes also express interest in the file with GET requests to 254 the sender, that file may later be PUT to a multicast address. 256 (Note that in the alternative Bundle Protocol, the Bundle Endpoint 257 Identifier (EID) can identify a group of endpoints, rather than just 258 one; mapping the Bundle EID onto multicast IP adddresses on IP 259 subnets is possible. Placing textual EIDs directly in HTTP-DTN's 260 Content-Source: and Content-Destination: headers, or in a Host: 261 field, would be possible to interwork HTTP-DTN and bundling.) 263 The utility of HTTP with multicast has been recognised previously as 264 a method of simple service discovery later adopted for the universal 265 plug and play (UPnP) protocol [I-D.draft-goland-http-udp] 266 [I-D.draft-cai-ssdp-v1]. Rather than call out multicast and unicast 267 separately as different protocols to be used by HTTP, recognising 268 that a given destination or address indicates multicast or broadcast 269 use should suffice. 271 Many existing HTTP/1.1 headers are directly useful with HTTP-DTN. 272 For example, ETag: headers are useful for identifying unique copies 273 of files in the network, and can be used to provide globally unique 274 identifiers (GUIDs) for each version of a file. Age: headers are 275 useful for estimating the amount of time a MIME object has been in 276 the network - indicating both transmission and storage times. Last- 277 Modified: times refer to the times on the origin server - that is, 278 the Content-Source: - and should be preserved during onward 279 forwarding. Max-Forwards: provides a TTL hop count and propagation 280 limitation mechanism. 282 3. Other useful proposed additional HTTP headers 284 A number of other additional HTTP headers are proposed here, as 285 likely to be useful. These SHOULD be implemented. These would 286 benefit from being specified more completely, in line with the 287 suggestions in [RFC2774]. 289 An HTTP object is just one binary file; the ability to group objects 290 together is useful (and is done in bundles by the Bundle Protocol). 291 If we call a group of related objects sent from the same source to 292 the same destination a 'package' (a name chosen to avoid any 293 confusion with the Bundle Protocol specification), we can then define 294 simple headers to be sent before each object: 296 Package-ID: - provides a unique textual identifier for the package 298 Package-Item: n of m (e.g. 1 of 7) - order of this HTTP file in the 299 package 301 Package-MD5: - MD5 hash across all Content-MD5: headers added 302 together in order of Package-Item: precedence. 304 A way to request missing Package-Items (from the previous node or 305 from the source) is likely to be very useful. 307 Precedence: headers could set importance of objects - very-high, 308 high, normal low, very-low - to give simple quality of service and 309 prioritization. 311 Some sort of header protection may be a good idea; Content-MD5: 312 covers the message body (entity-body), but not the headers. Header- 313 MD5: could cover some important HTTP headers. Header-MD5 could be 314 preserved across hops if possible, avoiding unnecessary header 315 reordering. Changing timestamps would invalidate the Header-MD5: 316 end-to-end, however - this needs more thought, particularly on where 317 timestamps are placed in HTTP headers. 319 For larger files, stronger mechanisms than MD5 should be examined. 321 There may be a need to send HTTP-DTN transfers across paths that 322 include hops with unidirectional one-way links with no return path, 323 e.g. when a wireless sender knows that a receiver is available, but 324 cannot hear it. Using: Connection: cannot-hear-response could be 325 used across that hop to indicate that the sender cannot hear 326 receivers. 328 Timestamps and how they are handled needs to be examined here in 329 greater detail. HTTP has the same basic assumption as the Bundle 330 Protocol - that all nodes are expected to know the current UTC time. 332 4. Other suggestions on using MIME in DTN networks 334 x-application-dtn has previously been proposed as a MIMEtype 335 identifying Bundle Protocol bundles delivered by HTTP. This provides 336 a way to support Bundle Protocol implementations in an HTTP 337 infrastructure. 339 Moving HTTP transfers over DTN networks using the Bundle Protocol has 340 already been proposed [Ott06]. By changing how HTTP is used - hop- 341 by-hop rather than end-to-end - HTTP can be used directly in DTN 342 networks without using the Bundle Protocol at all. 344 HTTP is a popular way to carry MIME, but support for MIME exists in 345 other protocols, including email, SIP and BEEP. BEEP can be thought 346 of as a more formalised and exactly-specified replacement for HTTP 347 for machine-machine interaction - and this detailed formal 348 specification makes BEEP complex [RFC3080]. BEEP provides an 349 alternative to HTTP to support XML-RPC and SOAP. BEEP's 350 specification is formally intended to support multiple different 351 transports, but only TCP transport of BEEP has been agreed [RFC3081]. 353 HTTP's simplicity of use and popularity appear to be compelling 354 advantages over BEEP. 356 5. Security Considerations 358 Better-Than-Nothing Security [RFC5386][RFC5387] is likely to be 359 useful here for ad-hoc communications without the availability of an 360 existing authentication infrastructure. 362 Security considerations and detailed examination of HTTP over TLS 363 (HTTPS) [RFC2817][RFC2818] and secure HTTP [RFC2660] are required 364 here. 366 Many existing security mechanism for HTTP could be used unchanged for 367 HTTP-DTN, if local conditions permit and the supporting 368 infrastructure, e.g. DNS, is available. However, reusing these 369 directly protects a single-hop transfer between peer nodes. To 370 protect an end-to-end transfer, the security mechanisms would need to 371 be applied using the information used in the Content-Source: and 372 Content-Destination: headers, before applying the local security 373 mechanism for the first peer-peer HTTP transfer. 375 6. IANA Considerations 377 Despite the Content-* rule for rejecting unfamiliar headers that 378 separates HTTP-DTN peers from traditional HTTP servers, it may be 379 desirable to use a non-standard port for DTN HTTP use over IP, rather 380 than the well-known port 80. If so, such a port should be requested 381 from IANA. 383 It may be necessary to request a dedicated IPv4 all-hosts multicast 384 address and a dedicated IPv6 link-local multicast addresses for local 385 HTTP DTN use, if local HTTP multicast is considered a desirable 386 feature. 388 7. Acknowledgements 390 We thank Wes Eddy and Kevin Fall for their review comments. 392 Work on the Saratoga protocol inspired some of the concepts that are 393 reused here, and we thank everyone involved in Saratoga's development 394 and implementation. 396 8. References 397 8.1. Normative References 399 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 400 Requirement Levels", BCP 14, RFC 2119, March 1997. 402 [RFC2616] Fielding, R., Gettys, J., Mogul, J., Frystyk, H., 403 Masinter, L., Leach, P., and T. Berners-Lee, "Hypertext 404 Transfer Protocol -- HTTP/1.1", RFC 2616, June 1999. 406 [RFC2774] Nielsen, H., Leach, P., and S. Lawrence, "An HTTP 407 Extension Framework", RFC 2774, February 2000. 409 8.2. Informative References 411 [Deering98] 412 Deering, S., "Watching the Waist of the Protocol 413 Hourglass", keynote, IEEE International Conference on 414 Network Protocols (ICNP), Austin Texas, October 1998. 416 [I-D.draft-cai-ssdp-v1] 417 Goland, Y., Cai, T., Leach, P., Gu, Y., and S. Albright, 418 "Simple Service Discovery Protocol/1.0 Operating without 419 an Arbiter", draft-cai-ssdp-v1-03 (expired) , 420 October 1999. 422 [I-D.draft-goland-http-udp] 423 Goland, Y., "Multicast and Unicast UDP HTTP Messages", 424 draft-goland-http-udp-01 (expired) , November 1999. 426 [I-D.irtf-dtnrg-bundle-checksum] 427 Eddy, W., Wood, L., and W. Ivancic, "Reliability-only 428 Ciphersuites for the Bundle Protocol", 429 draft-irtf-dtnrg-bundle-checksum-08 (work in progress) , 430 November 2010. 432 [I-D.natarajan-http-over-sctp] 433 Natarajan, P., Amer, P., Leighton, J., and F. Baker, 434 "Using SCTP as a Transport Layer Protocol for HTTP", 435 draft-natarajan-http-over-sctp-02 (work in progress), 436 July 2009. 438 [I-D.wood-tae-specifying-uri-transports] 439 Wood, L., "Specifying transport mechanisms in Uniform 440 Resource Identifiers", 441 draft-wood-tae-specifying-uri-transports-08 (work in 442 progress) , May 2010. 444 [I-D.wood-tsvwg-saratoga] 445 Wood, L., McKim, J., Eddy, W., Ivancic, W., and C. 446 Jackson, "Saratoga: A Scalable File Transfer Protocol", 447 draft-wood-tsvwg-saratoga-06 (work in progress) , 448 September 2010. 450 [Ott06] Ott, J. and D. Kutscher, "Bundling the Web: HTTP over 451 DTN", WNEPT 2006 Workshop on Networking in Public 452 Transport, QShine Conference, Ontario, August 2006. 454 [RFC2045] Freed, N. and N. Borenstein, "Multipurpose Internet Mail 455 Extensions (MIME) Part One: Format of Internet Message 456 Bodies", RFC 2045, November 1996. 458 [RFC2660] Rescorla, E. and A. Schiffman, "The Secure HyperText 459 Transfer Protocol", RFC 2660, August 1999. 461 [RFC2817] Khare, R. and S. Lawrence, "Upgrading to TLS Within 462 HTTP/1.1", RFC 2817, May 2000. 464 [RFC2818] Rescorla, E., "HTTP Over TLS", RFC 2818, May 2000. 466 [RFC3080] Rose, M., "The Blocks Extensible Exchange Protocol Core", 467 RFC 3080, March 2001. 469 [RFC3081] Rose, M., "Mapping the BEEP Core onto TCP", RFC 3081, 470 March 2001. 472 [RFC4838] Cerf, V., Burleigh, S., Hooke, A., Torgerson, L., Durst, 473 R., Scott, K., Fall, K., and H. Weiss, "Delay-Tolerant 474 Networking Architecture", RFC 4838, April 2007. 476 [RFC5050] Scott, K. and S. Burleigh, "Bundle Protocol 477 Specification", RFC 5050, November 2007. 479 [RFC5386] Williams, N. and M. Richardson, "Better-Than-Nothing 480 Security: An Unauthenticated Mode of IPsec", RFC 5386, 481 November 2008. 483 [RFC5387] Touch, J., Black, D., and Y. Wang, "Problem and 484 Applicability Statement for Better-Than-Nothing Security 485 (BTNS)", RFC 5387, November 2008. 487 [Wood09a] Wood, L., Eddy, W., and P. Holliday, "A Bundle of 488 Problems", IEEE Aerospace Conference, Big Sky, Montana, 489 March 2009. 491 [Wood09b] Wood, L., Holliday, P., Floreani, D., and I. Psaras, 492 "Moving data in DTNs with HTTP and MIME: Making use of 493 HTTP for delay- and disruption-tolerant networks with 494 convergence layers", Workshop on the Emergence of Delay-/ 495 Disruption-Tolerant Networks (e-DTN 2009), St Petersburg, 496 Russia, October 2009. 498 Authors' Addresses 500 Lloyd Wood 501 Centre for Communication Systems Research, University of Surrey 502 Guildford, Surrey GU2 7XH 503 United Kingdom 505 Phone: +44-1483-689123 506 Email: L.Wood@surrey.ac.uk 508 Peter Holliday 509 Cisco Systems 510 Level 12 511 300 Adelaide Street 512 Brisbane, Queensland 4000 513 Australia 515 Phone: +61-2-6216-0604 516 Email: phollida@cisco.com