idnits 2.17.1 draft-irtf-panrg-questions-12.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 : ---------------------------------------------------------------------------- ** The document seems to lack a Security Considerations section. ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (25 January 2022) is 820 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- No issues found here. Summary: 2 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Path Aware Networking RG B. Trammell 3 Internet-Draft Google Switzerland GmbH 4 Intended status: Informational 25 January 2022 5 Expires: 29 July 2022 7 Current Open Questions in Path Aware Networking 8 draft-irtf-panrg-questions-12 10 Abstract 12 In contrast to the present Internet architecture, a path-aware 13 internetworking architecture has two important properties: it exposes 14 the properties of available Internet paths to endpoints, and provides 15 for endpoints and applications to use these properties to select 16 paths through the Internet for their traffic. While this property of 17 "path awareness" already exists in many Internet-connected networks 18 within single domains and via administrative interfaces to the 19 network layer, a fully path-aware internetwork expands these concepts 20 across layers and across the Internet. 22 This document poses questions in path-aware networking open as of 23 2021, that must be answered in the design, development, and 24 deployment of path-aware internetworks. It was originally written to 25 frame discussions in the Path Aware Networking proposed Research 26 Group (PANRG), and has been published to snapshot current thinking in 27 this space. 29 Discussion Venues 31 This note is to be removed before publishing as an RFC. 33 Source for this draft and an issue tracker can be found at 34 https://github.com/panrg/questions. 36 Status of This Memo 38 This Internet-Draft is submitted in full conformance with the 39 provisions of BCP 78 and BCP 79. 41 Internet-Drafts are working documents of the Internet Engineering 42 Task Force (IETF). Note that other groups may also distribute 43 working documents as Internet-Drafts. The list of current Internet- 44 Drafts is at https://datatracker.ietf.org/drafts/current/. 46 Internet-Drafts are draft documents valid for a maximum of six months 47 and may be updated, replaced, or obsoleted by other documents at any 48 time. It is inappropriate to use Internet-Drafts as reference 49 material or to cite them other than as "work in progress." 51 This Internet-Draft will expire on 29 July 2022. 53 Copyright Notice 55 Copyright (c) 2022 IETF Trust and the persons identified as the 56 document authors. All rights reserved. 58 This document is subject to BCP 78 and the IETF Trust's Legal 59 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 60 license-info) in effect on the date of publication of this document. 61 Please review these documents carefully, as they describe your rights 62 and restrictions with respect to this document. Code Components 63 extracted from this document must include Revised BSD License text as 64 described in Section 4.e of the Trust Legal Provisions and are 65 provided without warranty as described in the Revised BSD License. 67 Table of Contents 69 1. Introduction to Path-Aware Networking . . . . . . . . . . . . 2 70 1.1. Definitions . . . . . . . . . . . . . . . . . . . . . . . 4 71 2. Questions . . . . . . . . . . . . . . . . . . . . . . . . . . 4 72 2.1. A Vocabulary of Path Properties . . . . . . . . . . . . . 5 73 2.2. Discovery, Distribution, and Trustworthiness of Path 74 Properties . . . . . . . . . . . . . . . . . . . . . . . 5 75 2.3. Supporting Path Selection . . . . . . . . . . . . . . . . 6 76 2.4. Interfaces for Path Awareness . . . . . . . . . . . . . . 6 77 2.5. Implications of Path Awareness for the Transport and 78 Application Layers . . . . . . . . . . . . . . . . . . . 7 79 2.6. What is an Endpoint? . . . . . . . . . . . . . . . . . . 7 80 2.7. Operating a Path Aware Network . . . . . . . . . . . . . 8 81 2.8. Deploying a Path Aware Network . . . . . . . . . . . . . 8 82 3. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 9 83 4. Informative References . . . . . . . . . . . . . . . . . . . 10 84 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 10 86 1. Introduction to Path-Aware Networking 88 In the current Internet architecture, the network layer provides a 89 best-effort service to the endpoints using it, without verifiability 90 of the properties of the path between tne endpoints. While there are 91 network layer technologies that attempt better-than-best-effort 92 delivery, the interfaces to these are generally administrative as 93 opposed to endpoint-exposed (e.g. Path Computation Element (PCE) 95 [RFC4655] and Software-Defined Wide Area Network (SD-WAN) 96 approaches), and they are often restricted to single administrative 97 domains. In this architecture, an application can assume that a 98 packet with a given destination address will eventually be forwarded 99 toward that destination, but little else. 101 A transport layer protocol such as TCP can provide reliability over 102 this best-effort service, and a protocol above the network layer, 103 such as Transport Layer Security (TLS) [RFC8446] can authenticate the 104 remote endpoint. However, little, if any, explicit information about 105 the path is available to the endpoints, and any assumptions made 106 about that path often do not hold. These sometimes have serious 107 impacts on the application, as in the case with BGP hijacking 108 attacks. 110 By contrast, in a path-aware internetworking architecture, endpoints 111 can select or influence the path(s) through the network used by any 112 given packet or flow. The network and transport layers explicitly 113 expose information about the path or paths available to the endpoints 114 and to the applications running on them, so that they can make this 115 selection. The Application Layer Traffic Optimization (ALTO) 116 protocol [RFC7285] can be seen as an example of a path-awareness 117 approach implemented in transport-layer terms on the present Internet 118 protocol stack. 120 Path selection provides explicit visibility and control of network 121 treatment to applications and users of the network. This selection 122 is available to the application, transport, and/or network layer 123 entities at each endpoint. Path control at the flow and subflow 124 level enables the design of new transport protocols that can leverage 125 multipath connectivity across disjoint paths through the Internet, 126 even over a single physical interface. When exposed to applications, 127 or to end-users through a system configuration interface, path 128 control allows the specification of constraints on the paths that 129 traffic should traverse, for instance to confound passive 130 surveillance in the network core [RFC7624]. 132 We note that this property of "path awareness" already exists in many 133 Internet-connected networks within single domains. Indeed, much of 134 the practice of network engineering using encapsulation at layer 3 135 can be said to be "path aware", in that it explicitly assigns traffic 136 at tunnel endpoints to a given path within the network. Path-aware 137 internetworking seeks to extend this awareness across domain 138 boundaries without resorting to overlays, except as a transition 139 technology. 141 This document presents a snapshot of open questions in this space 142 that will need to be answered in order to realize a path-aware 143 internetworking architecture; it is published to further frame 144 discussions within and outside the Path Aware Networking Research 145 Group, and is published with the rough consensus of that group. 147 1.1. Definitions 149 For purposes of this document, "path aware networking" describes 150 endpoint discovery of the properties of paths they use for 151 communication across an internetwork, and endpoint reaction to these 152 properties that affects routing and/or data transfer. Note that this 153 can and already does happen to some extent in the current Internet 154 architecture; this definition expands current techniques of path 155 discovery and manipulation to cross administrative domain boundaries 156 and up to the transport and application layers at the endpoints. 158 Expanding on this definition, a "path aware internetwork" is one in 159 which endpoint discovery of path properties and endpoint selection of 160 paths used by traffic exchanged by the endpoint are explicitly 161 supported, regardless of the specific design of the protocol features 162 which enable this discovery and selection. 164 A "path", for the purposes of these definitions, is abstractly 165 defined as a sequence of adjacent path elements over which a packet 166 can be transmitted, where the definition of "path element" is 167 technology-dependent. As this document is intended to pose questions 168 rather than answer them, it assumes that this definition will be 169 refined as part of the answer the first two questions it poses, about 170 the vocabulary of path properties and how they are disseminated. 172 Research into path aware internetworking covers any and all aspects 173 of designing, building, and operating path aware internetworks or the 174 networks and endpoints attached to them. This document presents a 175 collection of research questions to address in order to make a path 176 aware Internet a reality. 178 2. Questions 180 Realizing path-aware networking requires answers to a set of open 181 research questions. This document poses these questions, as a 182 starting point for discussions about how to realize path awareness in 183 the Internet, and to direct future research efforts within the Path 184 Aware Networking Research Group. 186 2.1. A Vocabulary of Path Properties 188 The first question: how are paths and path properties defined and 189 represented? 191 In order for information about paths to be exposed to an endpoint, 192 and for the endpoint to make use of that information, it is necessary 193 to define a common vocabulary for paths through an internetwork, and 194 properties of those paths. The elements of this vocabulary could 195 include terminology for components of a path and properties defined 196 for these components, for the entire path, or for subpaths of a path. 197 These properties may be relatively static, such as the presence of a 198 given node or service function on the path; as well as relatively 199 dynamic, such as the current values of metrics such as loss and 200 latency. 202 This vocabulary and its representation must be defined carefully, as 203 its design will have impacts on the properties (e.g., expressiveness, 204 scalability, security) of a given path-aware internetworking 205 architecture. For example, a system that exposes node-level 206 information for the topology through each network would maximize 207 information about the individual components of the path at the 208 endpoints, at the expense of making internal network topology 209 universally public, which may be in conflict with the business goals 210 of each network's operator. Furthermore, properties related to 211 individual components of the path may change frequently and may 212 quickly become outdated. However, aggregating the properties of 213 individual components to distill end-to-end properties for the entire 214 path is not trivial. 216 2.2. Discovery, Distribution, and Trustworthiness of Path Properties 218 The second question: how do endpoints and applications get access to 219 accurate, useful, and trustworthy path properties? 221 Once endpoints and networks have a shared vocabulary for expressing 222 path properties, the network must have some method for distributing 223 those path properties to the endpoints. Regardless of how path 224 property information is distributed, the endpoints require a method 225 to authenticate the properties -- to determine that they originated 226 from and pertain to the path that they purport to. 228 Choices in distribution and authentication methods will have impacts 229 on the scalability of a path-aware architecture. Possible dimensions 230 in the space of distribution methods include in-band versus out-of- 231 band, push versus pull versus publish-subscribe, and so on. There 232 are temporal issues with path property dissemination as well, 233 especially with dynamic properties, since the measurement or 234 elicitation of dynamic properties may be outdated by the time that 235 information is available at the endpoints, and interactions between 236 the measurement and dissemination delay may exhibit pathological 237 behavior for unlucky points in the parameter space. 239 2.3. Supporting Path Selection 241 The third question: how can endpoints select paths to use for traffic 242 in a way that can be trusted by the network, the endpoints, and the 243 applications using them? 245 Access to trustworthy path properties is only half of the challenge 246 in establishing a path-aware architecture. Endpoints must be able to 247 use this information in order to select paths for specific traffic 248 they send. As with the dissemination of path properties, choices 249 made in path selection methods will also have an impact on the 250 tradeoff between scalability and expressiveness of a path-aware 251 architecture. One key choice here is between in-band and out-of-band 252 control of path selection. Another is granularity of path selection 253 (whether per packet, per flow, or per larger aggregate), which also 254 has a large impact on the scalabilty/expressiveness tradeoff. Path 255 selection must, like path property information, be trustworthy, such 256 that the result of a path selection at an endpoint is predictable. 257 Moreover, any path selection mechanism should aim to provide an 258 outcome that is not worse than using a single path, or selecting 259 paths at random. 261 Path selection may be exposed in terms of the properties of the path 262 or the identity of elements of the path. In the latter case, a path 263 may be identified at any of multiple layers (e.g. routing domain 264 identifier, network layer address, higher-layer identifier or name, 265 and so on). In this case, care must be taken to present semantically 266 useful information to those making decisions about which path(s) to 267 trust. 269 2.4. Interfaces for Path Awareness 271 The fourth question: how can interfaces among the network, transport, 272 and application layers support the use of path awareness? 274 In order for applications to make effective use of a path-aware 275 networking architecture, the control interfaces presented by the 276 network and transport layers must also expose path properties to the 277 application in a useful way, and provide a useful set of paths among 278 which the application can select. Path selection must be possible 279 based not only on the preferences and policies of the application 280 developer, but of end-users as well. Also, the path selection 281 interfaces presented to applications and end users will need to 282 support multiple levels of granularity. Most applications' 283 requirements can be satisfied with the expression of path selection 284 policies in terms of properties of the paths, while some applications 285 may need finer-grained, per-path control. These interfaces will need 286 to support incremental development and deployment of applications, 287 and provide sensible defaults, to avoid hindering their adoption. 289 2.5. Implications of Path Awareness for the Transport and Application 290 Layers 292 The fifth question: how should transport-layer and higher layer 293 protocols be redesigned to work most effectively over a path-aware 294 networking layer? 296 In the current Internet, the basic assumption that at a given time 297 all traffic for a given flow will receive the same network treatment 298 and traverse the same path or equivalend paths often holds. In a 299 path aware network, this assumption is more easily violated. The 300 weakening of this assumption has implications for the design of 301 protocols above any path-aware network layer. 303 For example, one advantage of multipath communication is that a given 304 end-to-end flow can be "sprayed" along multiple paths in order to 305 confound attempts to collect data or metadata from those flows for 306 pervasive surveillance purposes [RFC7624]. However, the benefits of 307 this approach are reduced if the upper-layer protocols use linkable 308 identifiers on packets belonging to the same flow across different 309 paths. Clients may mitigate linkability by opting to not re-use 310 cleartext connection identifiers, such as TLS session IDs or tickets, 311 on separate paths. The privacy-conscious strategies required for 312 effective privacy in a path-aware Internet are only possible if 313 higher-layer protocols such as TLS permit clients to obtain 314 unlinkable identifiers. 316 2.6. What is an Endpoint? 318 The sixth question: how is path awareness (in terms of vocabulary and 319 interfaces) different when applied to tunnel and overlay endpoints? 321 The vision of path-aware networking articulated so far makes an 322 assumption that path properties will be disseminated to endpoints on 323 which applications are running (terminals with user agents, servers, 324 and so on). However, incremental deployment may require that a path- 325 aware network "core" be used to interconnect islands of legacy 326 protocol networks. In these cases, it is the gateways, not the 327 application endpoints, that receive path properties and make path 328 selections for that traffic. The interfaces provided by this gateway 329 are necessarily different than those a path-aware networking layer 330 provides to its transport and application layers, and the path 331 property information the gateway needs and makes available over those 332 interfaces may also be different. 334 2.7. Operating a Path Aware Network 336 The seventh question: how can a path aware network in a path aware 337 internetwork be effectively operated, given control inputs from 338 network administrators, application designers, and end users? 340 The network operations model in the current Internet architecture 341 assumes that traffic flows are controlled by the decisions and 342 policies made by network operators, as expressed in interdomain and 343 intradomain routing protocols. In a network providing path selection 344 to the endpoints, however, this assumption no longer holds, as 345 endpoints may react to path properties by selecting alternate paths. 346 Competing control inputs from path-aware endpoints and the routing 347 control plane may lead to more difficult traffic engineering or 348 nonconvergent forwarding, especially if the endpoints' and operators' 349 notion of the "best" path for given traffic diverges significantly. 350 The degree of difficulty may depend on the fidelity of information 351 made available to path selection algorithms at the endpoints. 352 Explicit path selection can also specify outbound paths, while BGP 353 policies are expressed in terms of inbound traffic. 355 A concept for path aware network operations will need to have clear 356 methods for the resolution of apparent (if not actual) conflicts of 357 intent between the network's operator and the path selection at an 358 endpoint. It will also need set of safety principles to ensure that 359 increasing path control does not lead to decreasing connectivity; one 360 such safety principle could be "the existence of at least one path 361 between two endpoints guarantees the selection of at least one path 362 between those endpoints." 364 2.8. Deploying a Path Aware Network 366 The eighth question: how can the incentives of network operators and 367 end-users be aligned to realize the vision of path aware networking, 368 and how can the transition from current ("path-oblivious") to path- 369 aware networking be managed? 371 The vision presented in the introduction discusses path aware 372 networking from the point of view of the benefits accruing at the 373 endpoints, to designers of transport protocols and applications as 374 well as to the end users of those applications. However, this vision 375 requires action not only at the endpoints but also within the 376 interconnected networks offering path aware connectivity. While the 377 specific actions required are a matter of the design and 378 implementation of a specific realization of a path aware protocol 379 stack, it is clear than any path aware architecture will require 380 network operators to give up some control of their networks over to 381 endpoint-driven control inputs. 383 Here the question of apparent versus actual conflicts of intent 384 arises again: certain network operations requirements may appear 385 essential, but are merely accidents of the interfaces provided by 386 current routing and management protocols. For example, related (but 387 adjacent) to path aware networking, the widespread use of the TCP 388 wire image [RFC8546] in network monitoring for DDoS prevention 389 appears in conflict with the deployment of encrypted transports, only 390 because path signaling [RFC8558] has been implicit in the deployment 391 of past transport protocols. 393 Similarly, incentives for deployment must show how existing network 394 operations requirements are met through new path selection and 395 property dissemination mechanisms. 397 The incentives for network operators and equipment vendors need to be 398 made clear, in terms of a plan to transition [RFC8170] an 399 internetwork to path-aware operation, one network and facility at a 400 time. This plan to transition must also take into account that the 401 dynamics of path aware networking early in this transition (when few 402 endpoints and flows in the Internet use path selection) may be 403 different than those later in the transition. 405 Aspects of data security and information management in a network that 406 explicitly radiates more information about the network's deployment 407 and configuration, and implicitly radiates information about endpoint 408 configuration and preference through path selection, must also be 409 addressed. 411 3. Acknowledgments 413 Many thanks to Adrian Perrig, Jean-Pierre Smith, Mirja Kuehlewind, 414 Olivier Bonaventure, Martin Thomson, Shwetha Bhandari, Chris Wood, 415 Lee Howard, Mohamed Boucadair, Thorben Krueger, Gorry Fairhurst, 416 Spencer Dawkins, Reese Enghardt, Laurent Ciavaglia, Stephen Farrell, 417 and Richard Yang, for discussions leading to questions in this 418 document, and for feedback on the document itself. 420 This work is partially supported by the European Commission under 421 Horizon 2020 grant agreement no. 688421 Measurement and Architecture 422 for a Middleboxed Internet (MAMI), and by the Swiss State Secretariat 423 for Education, Research, and Innovation under contract no. 15.0268. 424 This support does not imply endorsement. 426 4. Informative References 428 [RFC4655] Farrel, A., Vasseur, J.-P., and J. Ash, "A Path 429 Computation Element (PCE)-Based Architecture", RFC 4655, 430 DOI 10.17487/RFC4655, August 2006, 431 . 433 [RFC7285] Alimi, R., Ed., Penno, R., Ed., Yang, Y., Ed., Kiesel, S., 434 Previdi, S., Roome, W., Shalunov, S., and R. Woundy, 435 "Application-Layer Traffic Optimization (ALTO) Protocol", 436 RFC 7285, DOI 10.17487/RFC7285, September 2014, 437 . 439 [RFC7624] Barnes, R., Schneier, B., Jennings, C., Hardie, T., 440 Trammell, B., Huitema, C., and D. Borkmann, 441 "Confidentiality in the Face of Pervasive Surveillance: A 442 Threat Model and Problem Statement", RFC 7624, 443 DOI 10.17487/RFC7624, August 2015, 444 . 446 [RFC8170] Thaler, D., Ed., "Planning for Protocol Adoption and 447 Subsequent Transitions", RFC 8170, DOI 10.17487/RFC8170, 448 May 2017, . 450 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 451 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 452 . 454 [RFC8546] Trammell, B. and M. Kuehlewind, "The Wire Image of a 455 Network Protocol", RFC 8546, DOI 10.17487/RFC8546, April 456 2019, . 458 [RFC8558] Hardie, T., Ed., "Transport Protocol Path Signals", 459 RFC 8558, DOI 10.17487/RFC8558, April 2019, 460 . 462 Author's Address 464 Brian Trammell 465 Google Switzerland GmbH 466 Gustav-Gull-Platz 1 467 CH- 8004 Zurich 468 Switzerland 470 Email: ietf@trammell.ch