idnits 2.17.1 draft-ietf-ippm-ioam-ipv6-options-05.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 date (February 21, 2021) is 1157 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-17) exists of draft-ietf-ippm-ioam-data-11 == Outdated reference: A later version (-11) exists of draft-ietf-ippm-ioam-direct-export-02 -- No information found for draft-ietf-6man-hbh-header-handling - is the name correct? Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 ippm S. Bhandari 3 Internet-Draft Thoughtspot 4 Intended status: Standards Track F. Brockners 5 Expires: August 25, 2021 C. Pignataro 6 Cisco 7 H. Gredler 8 RtBrick Inc. 9 J. Leddy 10 Comcast 11 S. Youell 12 JMPC 13 T. Mizrahi 14 Huawei Network.IO Innovation Lab 15 A. Kfir 16 B. Gafni 17 Mellanox Technologies, Inc. 18 P. Lapukhov 19 Facebook 20 M. Spiegel 21 Barefoot Networks, an Intel company 22 S. Krishnan 23 Kaloom 24 R. Asati 25 Cisco 26 M. Smith 27 February 21, 2021 29 In-situ OAM IPv6 Options 30 draft-ietf-ippm-ioam-ipv6-options-05 32 Abstract 34 In-situ Operations, Administration, and Maintenance (IOAM) records 35 operational and telemetry information in the packet while the packet 36 traverses a path between two points in the network. This document 37 outlines how IOAM data fields are encapsulated in IPv6. 39 Status of This Memo 41 This Internet-Draft is submitted in full conformance with the 42 provisions of BCP 78 and BCP 79. 44 Internet-Drafts are working documents of the Internet Engineering 45 Task Force (IETF). Note that other groups may also distribute 46 working documents as Internet-Drafts. The list of current Internet- 47 Drafts is at https://datatracker.ietf.org/drafts/current/. 49 Internet-Drafts are draft documents valid for a maximum of six months 50 and may be updated, replaced, or obsoleted by other documents at any 51 time. It is inappropriate to use Internet-Drafts as reference 52 material or to cite them other than as "work in progress." 54 This Internet-Draft will expire on August 25, 2021. 56 Copyright Notice 58 Copyright (c) 2021 IETF Trust and the persons identified as the 59 document authors. All rights reserved. 61 This document is subject to BCP 78 and the IETF Trust's Legal 62 Provisions Relating to IETF Documents 63 (https://trustee.ietf.org/license-info) in effect on the date of 64 publication of this document. Please review these documents 65 carefully, as they describe your rights and restrictions with respect 66 to this document. Code Components extracted from this document must 67 include Simplified BSD License text as described in Section 4.e of 68 the Trust Legal Provisions and are provided without warranty as 69 described in the Simplified BSD License. 71 Table of Contents 73 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 74 2. Conventions . . . . . . . . . . . . . . . . . . . . . . . . . 3 75 2.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 76 2.2. Abbreviations . . . . . . . . . . . . . . . . . . . . . . 3 77 3. In-situ OAM Metadata Transport in IPv6 . . . . . . . . . . . 3 78 4. IOAM Deployment In IPv6 Networks . . . . . . . . . . . . . . 6 79 4.1. Considerations for IOAM deployment in IPv6 networks . . . 6 80 4.2. IOAM domains bounded by hosts . . . . . . . . . . . . . . 7 81 4.3. IOAM domains bounded by network devices . . . . . . . . . 7 82 4.4. Deployment options . . . . . . . . . . . . . . . . . . . 8 83 4.4.1. IPv6-in-IPv6 encapsulation . . . . . . . . . . . . . 8 84 4.4.2. IP-in-IPv6 encapsulation with ULA . . . . . . . . . . 8 85 4.4.3. x-in-IPv6 Encapsulation that is used Independently . 9 86 5. Security Considerations . . . . . . . . . . . . . . . . . . . 9 87 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 88 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 89 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 90 8.1. Normative References . . . . . . . . . . . . . . . . . . 10 91 8.2. Informative References . . . . . . . . . . . . . . . . . 11 92 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 11 94 1. Introduction 96 In-situ Operations, Administration, and Maintenance (IOAM) records 97 operational and telemetry information in the packet while the packet 98 traverses a path between two points in the network. This document 99 outlines how IOAM data fields are encapsulated in the IPv6 [RFC8200] 100 and discusses deployment options for networks that use 101 IPv6-encapsulated IOAM data fields. These options have distinct 102 deployment considerations; for example, the IOAM domain can either be 103 between hosts, or be between IOAM encapsulating and decapsulating 104 network nodes that forward traffic, such as routers. 106 2. Conventions 108 2.1. Requirements Language 110 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 111 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 112 "OPTIONAL" in this document are to be interpreted as described in BCP 113 14 [RFC2119] [RFC8174] when, and only when, they appear in all 114 capitals, as shown here. 116 2.2. Abbreviations 118 Abbreviations used in this document: 120 E2E: Edge-to-Edge 122 IOAM: In-situ Operations, Administration, and Maintenance 124 ION: IOAM Overlay Network 126 OAM: Operations, Administration, and Maintenance 128 POT: Proof of Transit 130 3. In-situ OAM Metadata Transport in IPv6 132 In-situ OAM in IPv6 is used to enhance diagnostics of IPv6 networks. 133 It complements other mechanisms designed to enhance diagnostics of 134 IPv6 networks, such as the IPv6 Performance and Diagnostic Metrics 135 Destination Option described in [RFC8250]. 137 IOAM data fields can be encapsulated in "option data" fields using 138 two types of extension headers in IPv6 packets - either Hop-by-Hop 139 Options header or Destination options header. Deployments select one 140 of these extension header types depending on how IOAM is used, as 141 described in section 4 of [I-D.ietf-ippm-ioam-data]. Multiple 142 options with the same Option Type MAY appear in the same Hop-by-Hop 143 Options or Destination Options header, with distinct content. 145 In order for IOAM to work in IPv6 networks, IOAM MUST be explicitly 146 enabled per interface on every node within the IOAM domain. Unless a 147 particular interface is explicitly enabled (i.e., explicitly 148 configured) for IOAM, a router MUST drop packets that contain 149 extension headers carrying IOAM data-fields. This is the default 150 behavior and is independent of whether the Hop-by-Hop options or 151 Destination options are used to encode the IOAM data. This ensures 152 that IOAM data does not unintentionally get forwarded outside the 153 IOAM domain. 155 An IPv6 packet carrying IOAM data in an Extension header can have 156 other extension headers, compliant with [RFC8200]. 158 IPv6 Hop-by-Hop and Destination Option format for carrying in-situ 159 OAM data fields: 161 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 162 | Option Type | Opt Data Len | Reserved | IOAM Type | 163 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<-+ 164 | | | 165 . . I 166 . . O 167 . . A 168 . . M 169 . . . 170 . Option Data . O 171 . . P 172 . . T 173 . . I 174 . . O 175 . . N 176 | | | 177 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<-+ 179 Option Type: 8-bit option type identifier as defined inSection 6. 181 Opt Data Len: 8-bit unsigned integer. Length of this option, in 182 octets, not including the first 2 octets. 184 Reserved: 8-bit field MUST be set to zero upon transmission and 185 ignored upon reception. 187 IOAM Type: 8-bit field as defined in section 7.2 in 188 [I-D.ietf-ippm-ioam-data]. 190 Option Data: Variable-length field. Option-Type-specific data. 192 In-situ OAM Option-Types are inserted as Option data as follows: 194 1. Pre-allocated Trace Option: The in-situ OAM Preallocated Trace 195 Option-Type defined in [I-D.ietf-ippm-ioam-data] is represented 196 as an IPv6 option in the Hop-by-Hop extension header: 198 Option Type: 001xxxxx 8-bit identifier of the IOAM type of 199 option. xxxxx=TBD. 201 IOAM Option-Type: IOAM Pre-allocated Trace Option-Type. 203 2. Incremental Trace Option: The in-situ OAM Incremental Trace 204 Option-Type defined in [I-D.ietf-ippm-ioam-data] is represented 205 as an IPv6 option in the Hop-by-Hop extension header: 207 Option Type: 001xxxxx 8-bit identifier of the IOAM type of 208 option. xxxxx=TBD. 210 IOAM Option-Type: IOAM Incremental Trace Option-Type. 212 3. Proof of Transit Option: The in-situ OAM POT Option-Type defined 213 in [I-D.ietf-ippm-ioam-data] is represented as an IPv6 option in 214 the Hop-by-Hop extension header: 216 Option Type: 001xxxxx 8-bit identifier of the IOAM type of 217 option. xxxxx=TBD. 219 IOAM Option-Type: IOAM POT Option-Type. 221 4. Edge to Edge Option: The in-situ OAM E2E option defined in 222 [I-D.ietf-ippm-ioam-data] is represented as an IPv6 option in 223 Destination extension header: 225 Option Type: 000xxxxx 8-bit identifier of the IOAM type of 226 option. xxxxx=TBD. 228 IOAM Option-Type: IOAM E2E Option-Type. 230 5. Direct Export (DEX) Option: The in-situ OAM Direct Export Option- 231 Type defined in [I-D.ietf-ippm-ioam-direct-export] is represented 232 as an IPv6 option in the Hop-by-Hop extension header: 234 Option Type: 000xxxxx 8-bit identifier of the IOAM type of 235 option. xxxxx=TBD. 237 IOAM Option-Type: IOAM Direct Export (DEX) Option-Type. 239 All the in-situ OAM IPv6 options defined here have alignment 240 requirements. Specifically, they all require 4n alignment. This 241 ensures that fields specified in [I-D.ietf-ippm-ioam-data] are 242 aligned at a multiple-of-4 offset from the start of the Hop-by-Hop 243 and Destination Options header. In addition, to maintain IPv6 244 extension header 8-octet alignment and avoid the need to add or 245 remove padding at every hop, the Trace-Type for Incremental Trace 246 Option in IPv6 MUST be selected such that the IOAM node data length 247 is a multiple of 8-octets. 249 IPv6 options can have a maximum length of 255 octets. Consequently, 250 the total lenght of IOAM Option-Types including all data fields is 251 also limited to 255 octets when encapsulated into IPv6. 253 4. IOAM Deployment In IPv6 Networks 255 4.1. Considerations for IOAM deployment in IPv6 networks 257 IOAM deployments in IPv6 networks should take the following 258 considerations and requirements into account: 260 C1 It is desirable that the addition of IOAM data fields neither 261 changes the way routers forward packets nor the forwarding 262 decisions the routers take. Packets with added OAM information 263 should follow the same path within the domain that an identical 264 packet without OAM information would follow, even in the presence 265 of ECMP. Such behavior is particularly important for deployments 266 where IOAM data fields are only added "on-demand", e.g., to 267 provide further insights in case of undesired network behavior for 268 certain flows. Implementations of IOAM SHOULD ensure that ECMP 269 behavior for packets with and without IOAM data fields is the 270 same. 272 C2 Given that IOAM data fields increase the total size of a packet, 273 the size of a packet including the IOAM data could exceed the 274 PMTU. In particular, the incremental trace IOAM Hop-by-Hop (HbH) 275 Option, which is intended to support hardware implementations of 276 IOAM, changes Option Data Length en-route. Operators of an IOAM 277 domain SHOULD ensure that the addition of OAM information does not 278 lead to fragmentation of the packet, e.g., by configuring the MTU 279 of transit routers and switches to a sufficiently high value. 280 Careful control of the MTU in a network is one of the reasons why 281 IOAM is considered a domain-specific feature (see also 283 [I-D.ietf-ippm-ioam-data]). In addition, the PMTU tolerance range 284 in the IOAM domain should be identified (e.g., through 285 configuration) and IOAM encapsulation operations and/or IOAM data 286 field insertion (in case of incremental tracing) should not be 287 performed if it exceeds the packet size beyond PMTU. 289 C3 Packets with IOAM data or associated ICMP errors, should not 290 arrive at destinations that have no knowledge of IOAM. For 291 exmample, if IOAM is used in in transit devices, misleading ICMP 292 errors due to addition and/or presence of OAM data in a packet 293 could confuse the host that sent the packet if it did not insert 294 the OAM information. 296 C4 OAM data leaks can affect the forwarding behavior and state of 297 network elements outside an IOAM domain. IOAM domains SHOULD 298 provide a mechanism to prevent data leaks or be able to ensure 299 that if a leak occurs, network elements outside the domain are not 300 affected (i.e., they continue to process other valid packets). 302 C5 The source that inserts and leaks the IOAM data needs to be easy 303 to identify for the purpose of troubleshooting, due to the high 304 complexity of troubleshooting a source that inserted the IOAM data 305 and did not remove it when the packet traversed across an 306 Autonomous System (AS). Such a troubleshooting process might 307 require coordination between multiple operators, complex 308 configuration verification, packet capture analysis, etc. 310 C6 Compliance with [RFC8200] requires OAM data to be encapsulated 311 instead of header/option insertion directly into in-flight packets 312 using the original IPv6 header. 314 4.2. IOAM domains bounded by hosts 316 For deployments where the IOAM domain is bounded by hosts, hosts will 317 perform the operation of IOAM data field encapsulation and 318 decapsulation. IOAM data is carried in IPv6 packets as Hop-by-Hop or 319 Destination options as specified in this document. 321 4.3. IOAM domains bounded by network devices 323 For deployments where the IOAM domain is bounded by network devices, 324 network devices such as routers form the edge of an IOAM domain. 325 Network devices will perform the operation of IOAM data field 326 encapsulation and decapsulation. 328 4.4. Deployment options 330 This section lists out possible deployment options that can be 331 employed to meet the requirements listed in Section 4.1. 333 4.4.1. IPv6-in-IPv6 encapsulation 335 The "IPv6-in-IPv6" approach preserves the original IP packet and add 336 an IPv6 header including IOAM data fields in an extension header in 337 front of it, to forward traffic within and across an IOAM domain. 338 The overlay network formed by the additional IPv6 header with the 339 IOAM data fields included in an extension header is referred to as 340 IOAM Overlay Network (ION) in this document. 342 The following steps should be taken to perform an IPv6-in-IPv6 343 approach: 345 1. The source address of the outer IPv6 header is that of the IOAM 346 encapsulating node. The destination address of the outer IPv6 347 header is the same as the inner IPv6 destination address, i.e., 348 the destination address of the packet does not change. 350 2. To simplify debugging in case of leaked IOAM data fields, 351 consider a new IOAM E2E destination option to identify the Source 352 IOAM domain (AS, v6 prefix). Insert this option into the IOAM 353 destination options EH attached to the outer IPv6 header. This 354 additional information would allow for easy identification of an 355 AS operator that is the source of packets with leaked IOAM 356 information. Note that leaked packets with IOAM data fields 357 would only occur in case a router would be misconfigured. 359 3. All the IOAM options are defined with type "00" - skip over this 360 option and continue processing the header. Presence of these 361 options must not cause packet drops in network elements that do 362 not understand the option. In addition, 363 [I-D.ietf-6man-hbh-header-handling] should be considered. 365 4.4.2. IP-in-IPv6 encapsulation with ULA 367 The "IP-in-IPv6 encapsulation with ULA" [RFC4193] approach can be 368 used to apply IOAM to either an IPv6 or an IPv4 network. In 369 addition, it fulfills requirement C4 (avoid leaks) by using ULA for 370 the ION. Similar to the IPv6-in-IPv6 encapsulation approach above, 371 the original IP packet is preserved. An IPv6 header including IOAM 372 data fields in an extension header is added in front of it, to 373 forward traffic within and across the IOAM domain. IPv6 addresses 374 for the ION, i.e. the outer IPv6 addresses are assigned from the ULA 375 space. Addressing and routing in the ION are to be configured so 376 that the IP-in-IPv6 encapsulated packets follow the same path as the 377 original, non-encapsulated packet would have taken. This would 378 create an internal IPv6 forwarding topology using the IOAM domain's 379 interior ULA address space which is parallel with the forwarding 380 topology that exists with the non-IOAM address space (the topology 381 and address space that would be followed by packets that do not have 382 supplemental IOAM information). Establishment and maintenance of the 383 parallel IOAM ULA forwarding topology could be automated, e.g., 384 similar to how LDP [RFC5036] is used in MPLS to establish and 385 maintain an LSP forwarding topology that is parallel to the network's 386 IGP forwarding topology. 388 Transit across the ION could leverage the transit approach for 389 traffic between BGP border routers, as described in [RFC1772], "A.2.3 390 Encapsulation". Assuming that the operational guidelines specified 391 in Section 4 of [RFC4193] are properly followed, the probability of 392 leaks in this approach will be almost close to zero. If the packets 393 do leak through IOAM egress device misconfiguration or partial IOAM 394 egress device failure, the packets' ULA destination address is 395 invalid outside of the IOAM domain. There is no exterior destination 396 to be reached, and the packets will be dropped when they encounter 397 either a router external to the IOAM domain that has a packet filter 398 that drops packets with ULA destinations, or a router that does not 399 have a default route. 401 4.4.3. x-in-IPv6 Encapsulation that is used Independently 403 In some cases it is desirable to monitor a domain that uses an 404 overlay network that is deployed independently of the need for IOAM, 405 e.g., an overlay network that runs Geneve-in-IPv6, or VXLAN-in-IPv6. 406 In this case IOAM can be encapsulated in as an extension header in 407 the tunnel (outer) IPv6 header. Thus, the tunnel encapsulating node 408 is also the IOAM encapsulating node, and the tunnel end point is also 409 the IOAM decapsulating node. 411 5. Security Considerations 413 This document describes the encapsulation of IOAM data fields in 414 IPv6. Security considerations of the specific IOAM data fields for 415 each case (i.e., Trace, Proof of Transit, and E2E) are described and 416 defined in [I-D.ietf-ippm-ioam-data]. 418 As this document describes new options for IPv6, these are similar to 419 the security considerations of [RFC8200] and the weakness documented 420 in [RFC8250]. 422 6. IANA Considerations 424 This draft requests the following IPv6 Option Type assignments from 425 the Destination Options and Hop-by-Hop Options sub-registry of 426 Internet Protocol Version 6 (IPv6) Parameters. 428 http://www.iana.org/assignments/ipv6-parameters/ipv6- 429 parameters.xhtml#ipv6-parameters-2 431 Hex Value Binary Value Description Reference 432 act chg rest 433 ---------------------------------------------------------------- 434 TBD_1_0 00 0 TBD_1 IOAM [This draft] 435 TBD_1_1 00 1 TBD_1 IOAM [This draft] 437 7. Acknowledgements 439 The authors would like to thank Tom Herbert, Eric Vyncke, Nalini 440 Elkins, Srihari Raghavan, Ranganathan T S, Karthik Babu Harichandra 441 Babu, Akshaya Nadahalli, Stefano Previdi, Hemant Singh, Erik 442 Nordmark, LJ Wobker, Mark Smith, Andrew Yourtchenko and Justin Iurman 443 for the comments and advice. For the IPv6 encapsulation, this 444 document leverages concepts described in 445 [I-D.kitamura-ipv6-record-route]. The authors would like to 446 acknowledge the work done by the author Hiroshi Kitamura and people 447 involved in writing it. 449 8. References 451 8.1. Normative References 453 [I-D.ietf-ippm-ioam-data] 454 Brockners, F., Bhandari, S., and T. Mizrahi, "Data Fields 455 for In-situ OAM", draft-ietf-ippm-ioam-data-11 (work in 456 progress), November 2020. 458 [I-D.ietf-ippm-ioam-direct-export] 459 Song, H., Gafni, B., Zhou, T., Li, Z., Brockners, F., 460 Bhandari, S., Sivakolundu, R., and T. Mizrahi, "In-situ 461 OAM Direct Exporting", draft-ietf-ippm-ioam-direct- 462 export-02 (work in progress), November 2020. 464 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 465 Requirement Levels", BCP 14, RFC 2119, 466 DOI 10.17487/RFC2119, March 1997, 467 . 469 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 470 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 471 May 2017, . 473 8.2. Informative References 475 [I-D.ietf-6man-hbh-header-handling] 476 Baker, F. and R. Bonica, "IPv6 Hop-by-Hop Options 477 Extension Header", March 2016. 479 [I-D.kitamura-ipv6-record-route] 480 Kitamura, H., "Record Route for IPv6 (PR6) Hop-by-Hop 481 Option Extension", draft-kitamura-ipv6-record-route-00 482 (work in progress), November 2000. 484 [RFC1772] Rekhter, Y. and P. Gross, "Application of the Border 485 Gateway Protocol in the Internet", RFC 1772, 486 DOI 10.17487/RFC1772, March 1995, 487 . 489 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 490 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 491 . 493 [RFC5036] Andersson, L., Ed., Minei, I., Ed., and B. Thomas, Ed., 494 "LDP Specification", RFC 5036, DOI 10.17487/RFC5036, 495 October 2007, . 497 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 498 (IPv6) Specification", STD 86, RFC 8200, 499 DOI 10.17487/RFC8200, July 2017, 500 . 502 [RFC8250] Elkins, N., Hamilton, R., and M. Ackermann, "IPv6 503 Performance and Diagnostic Metrics (PDM) Destination 504 Option", RFC 8250, DOI 10.17487/RFC8250, September 2017, 505 . 507 Authors' Addresses 509 Shwetha Bhandari 510 Thoughtspot 511 3rd Floor, Indiqube Orion, 24th Main Rd, Garden Layout, HSR Layout 512 Bangalore, KARNATAKA 560 102 513 India 515 Email: shwetha.bhandari@thoughtspot.com 516 Frank Brockners 517 Cisco Systems, Inc. 518 Kaiserswerther Str. 115, 519 RATINGEN, NORDRHEIN-WESTFALEN 40880 520 Germany 522 Email: fbrockne@cisco.com 524 Carlos Pignataro 525 Cisco Systems, Inc. 526 7200-11 Kit Creek Road 527 Research Triangle Park, NC 27709 528 United States 530 Email: cpignata@cisco.com 532 Hannes Gredler 533 RtBrick Inc. 535 Email: hannes@rtbrick.com 537 John Leddy 538 Comcast 540 Email: John_Leddy@cable.comcast.com 542 Stephen Youell 543 JP Morgan Chase 544 25 Bank Street 545 London E14 5JP 546 United Kingdom 548 Email: stephen.youell@jpmorgan.com 550 Tal Mizrahi 551 Huawei Network.IO Innovation Lab 552 Israel 554 Email: tal.mizrahi.phd@gmail.com 555 Aviv Kfir 556 Mellanox Technologies, Inc. 557 350 Oakmead Parkway, Suite 100 558 Sunnyvale, CA 94085 559 U.S.A. 561 Email: avivk@mellanox.com 563 Barak Gafni 564 Mellanox Technologies, Inc. 565 350 Oakmead Parkway, Suite 100 566 Sunnyvale, CA 94085 567 U.S.A. 569 Email: gbarak@mellanox.com 571 Petr Lapukhov 572 Facebook 573 1 Hacker Way 574 Menlo Park, CA 94025 575 US 577 Email: petr@fb.com 579 Mickey Spiegel 580 Barefoot Networks, an Intel company 581 4750 Patrick Henry Drive 582 Santa Clara, CA 95054 583 US 585 Email: mickey.spiegel@intel.com 587 Suresh Krishnan 588 Kaloom 590 Email: suresh@kaloom.com 591 Rajiv Asati 592 Cisco Systems, Inc. 593 7200 Kit Creek Road 594 Research Triangle Park, NC 27709 595 US 597 Email: rajiva@cisco.com 599 Mark Smith 600 PO BOX 521 601 HEIDELBERG, VIC 3084 602 AU 604 Email: markzzzsmith+id@gmail.com