idnits 2.17.1 draft-ietf-detnet-architecture-02.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 == Line 630 has weird spacing: '...e stack v ...' == Line 1132 has weird spacing: '...mediate inter...' == Line 1135 has weird spacing: '...mediate inter...' -- The document date (June 29, 2017) is 2486 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) == Missing Reference: 'Network' is mentioned on line 766, but not defined == Unused Reference: 'AVnu' is defined on line 1696, but no explicit reference was found in the text == Unused Reference: 'HART' is defined on line 1705, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-6tisch-architecture' is defined on line 1724, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-6tisch-tsch' is defined on line 1729, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-roll-rpl-industrial-applicability' is defined on line 1748, but no explicit reference was found in the text == Unused Reference: 'I-D.svshah-tsvwg-deterministic-forwarding' is defined on line 1754, but no explicit reference was found in the text == Unused Reference: 'IEEE802.11-2012' is defined on line 1764, but no explicit reference was found in the text == Unused Reference: 'IEEE802.1AS-2011' is defined on line 1770, but no explicit reference was found in the text == Unused Reference: 'IEEE802.1TSNTG' is defined on line 1814, but no explicit reference was found in the text == Unused Reference: 'IEEE802154' is defined on line 1828, but no explicit reference was found in the text == Unused Reference: 'IEEE802154e' is defined on line 1834, but no explicit reference was found in the text == Unused Reference: 'ISA95' is defined on line 1845, but no explicit reference was found in the text == Unused Reference: 'ODVA' is defined on line 1849, but no explicit reference was found in the text == Unused Reference: 'Profinet' is defined on line 1856, but no explicit reference was found in the text == Unused Reference: 'RFC5673' is defined on line 1912, but no explicit reference was found in the text == Unused Reference: 'WirelessHART' is defined on line 1945, but no explicit reference was found in the text == Outdated reference: A later version (-30) exists of draft-ietf-6tisch-architecture-11 == Outdated reference: A later version (-09) exists of draft-ietf-detnet-problem-statement-01 == Outdated reference: A later version (-20) exists of draft-ietf-detnet-use-cases-12 -- Obsolete informational reference (is this intentional?): RFC 5316 (Obsoleted by RFC 9346) Summary: 0 errors (**), 0 flaws (~~), 24 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DetNet N. Finn 3 Internet-Draft Huawei Technologies Co. Ltd 4 Intended status: Standards Track P. Thubert 5 Expires: December 31, 2017 Cisco 6 B. Varga 7 J. Farkas 8 Ericsson 9 June 29, 2017 11 Deterministic Networking Architecture 12 draft-ietf-detnet-architecture-02 14 Abstract 16 Deterministic Networking (DetNet) provides a capability to carry 17 specified unicast or multicast data flows for real-time applications 18 with extremely low data loss rates and bounded latency. Techniques 19 used include: 1) reserving data plane resources for individual (or 20 aggregated) DetNet flows in some or all of the intermediate nodes 21 (e.g. bridges or routers) along the path of the flow; 2) providing 22 explicit routes for DetNet flows that do not rapidly change with the 23 network topology; and 3) distributing data from DetNet flow packets 24 over time and/or space to ensure delivery of each packet's data' in 25 spite of the loss of a path. The capabilities can be managed by 26 configuration, or by manual or automatic network management. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at http://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on December 31, 2017. 45 Copyright Notice 47 Copyright (c) 2017 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (http://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 64 2.1. Terms used in this document . . . . . . . . . . . . . . . 4 65 2.2. IEEE 802 TSN to DetNet dictionary . . . . . . . . . . . . 6 66 3. Providing the DetNet Quality of Service . . . . . . . . . . . 7 67 3.1. Primary goals defining the DetNet QoS . . . . . . . . . . 7 68 3.2. Mechanisms to achieve DetNet Qos . . . . . . . . . . . . 9 69 3.2.1. Congestion protection . . . . . . . . . . . . . . . . 9 70 3.2.2. Explicit routes . . . . . . . . . . . . . . . . . . . 9 71 3.2.3. Jitter Reduction . . . . . . . . . . . . . . . . . . 10 72 3.2.4. Packet Replication and Elimination . . . . . . . . . 11 73 3.3. Secondary goals for DetNet . . . . . . . . . . . . . . . 12 74 3.3.1. Coexistence with normal traffic . . . . . . . . . . . 12 75 3.3.2. Fault Mitigation . . . . . . . . . . . . . . . . . . 13 76 4. DetNet Architecture . . . . . . . . . . . . . . . . . . . . . 14 77 4.1. DetNet stack model . . . . . . . . . . . . . . . . . . . 14 78 4.1.1. Representative Protocol Stack Model . . . . . . . . . 14 79 4.1.2. DetNet Data Plane Overview . . . . . . . . . . . . . 16 80 4.1.3. Network reference model . . . . . . . . . . . . . . . 18 81 4.2. DetNet systems . . . . . . . . . . . . . . . . . . . . . 19 82 4.2.1. End system . . . . . . . . . . . . . . . . . . . . . 19 83 4.2.2. DetNet edge, relay, and transit nodes . . . . . . . . 20 84 4.3. DetNet flows . . . . . . . . . . . . . . . . . . . . . . 21 85 4.3.1. DetNet flow types . . . . . . . . . . . . . . . . . . 21 86 4.3.2. Source guarantees . . . . . . . . . . . . . . . . . . 21 87 4.3.3. Incomplete Networks . . . . . . . . . . . . . . . . . 23 88 4.4. Traffic Engineering for DetNet . . . . . . . . . . . . . 23 89 4.4.1. The Application Plane . . . . . . . . . . . . . . . . 23 90 4.4.2. The Controller Plane . . . . . . . . . . . . . . . . 24 91 4.4.3. The Network Plane . . . . . . . . . . . . . . . . . . 24 92 4.5. Queuing, Shaping, Scheduling, and Preemption . . . . . . 25 93 4.6. Service instance . . . . . . . . . . . . . . . . . . . . 26 94 4.7. Flow identification at technology borders . . . . . . . . 27 95 4.7.1. Exporting flow identification . . . . . . . . . . . . 27 96 4.7.2. Flow attribute mapping between layers . . . . . . . . 29 97 4.7.3. Flow-ID mapping examples . . . . . . . . . . . . . . 30 98 4.8. Advertising resources, capabilities and adjacencies . . . 32 99 4.9. Provisioning model . . . . . . . . . . . . . . . . . . . 32 100 4.9.1. Centralized Path Computation and Installation . . . . 32 101 4.9.2. Distributed Path Setup . . . . . . . . . . . . . . . 32 102 4.10. Scaling to larger networks . . . . . . . . . . . . . . . 33 103 4.11. Connected islands vs. networks . . . . . . . . . . . . . 33 104 4.12. Compatibility with Layer-2 . . . . . . . . . . . . . . . 33 105 5. Open Questions . . . . . . . . . . . . . . . . . . . . . . . 34 106 5.1. Flat vs. hierarchical control . . . . . . . . . . . . . . 34 107 5.2. Peer-to-peer reservation protocol . . . . . . . . . . . . 34 108 5.3. Wireless media interactions . . . . . . . . . . . . . . . 35 109 5.4. Packet encoding for service protection . . . . . . . . . 35 110 6. Security Considerations . . . . . . . . . . . . . . . . . . . 35 111 7. Privacy Considerations . . . . . . . . . . . . . . . . . . . 36 112 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 36 113 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 36 114 10. Access to IEEE 802.1 documents . . . . . . . . . . . . . . . 37 115 11. Informative References . . . . . . . . . . . . . . . . . . . 37 116 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 42 118 1. Introduction 120 Deterministic Networking (DetNet) is a service that can be offered by 121 a network to DetNet flows. DetNet provides these flows extremely low 122 packet loss rates and assured maximum end-to-end delivery latency. 123 This is accomplished by dedicating network resources such as link 124 bandwidth and buffer space to DetNet flows and/or classes of DetNet 125 flows, and by replicating packets along multiple paths. Unused 126 reserved resources are available to non-DetNet packets. 128 The Deterministic Networking Problem Statement 129 [I-D.ietf-detnet-problem-statement] introduces Deterministic 130 Networking, and Deterministic Networking Use Cases 131 [I-D.ietf-detnet-use-cases] summarizes the need for it. See 132 [I-D.dt-detnet-dp-alt] for a discussion of specific techniques that 133 can be used to identify DetNet Flows and assign them to specific 134 paths through a network. 136 A goal of DetNet is a converged network in all respects. That is, 137 the presence of DetNet flows does not preclude non-DetNet flows, and 138 the benefits offered DetNet flows should not, except in extreme 139 cases, prevent existing QoS mechanisms from operating in a normal 140 fashion, subject to the bandwidth required for the DetNet flows. A 141 single source-destination pair can trade both DetNet and non-DetNet 142 flows. End systems and applications need not instantiate special 143 interfaces for DetNet flows. Networks are not restricted to certain 144 topologies; connectivity is not restricted. Any application that 145 generates a data flow that can be usefully characterized as having a 146 maximum bandwidth should be able to take advantage of DetNet, as long 147 as the necessary resources can be reserved. Reservations can be made 148 by the application itself, via network management, by an applications 149 controller, or by other means. 151 Many applications of interest to Deterministic Networking require the 152 ability to synchronize the clocks in end systems to a sub-microsecond 153 accuracy. Some of the queue control techniques defined in 154 Section 4.5 also require time synchronization among relay and transit 155 nodes. The means used to achieve time synchronization are not 156 addressed in this document. DetNet should accommodate various 157 synchronization techniques and profiles that are defined elsewhere to 158 solve exchange time in different market segments. 160 The present document is an individual contribution, but it is 161 intended by the authors for adoption by the DetNet working group. 163 2. Terminology 165 2.1. Terms used in this document 167 The following special terms are used in this document in order to 168 avoid the assumption that a given element in the architecture does or 169 does not have Internet Protocol stack, functions as a router, bridge, 170 firewall, or otherwise plays a particular role at Layer-2 or higher. 172 App-flow 173 The native format of a DetNet flow. 175 destination 176 An end system capable of receiving a DetNet flow. 178 DetNet domain 179 The portion of a network that is DetNet aware. It includes 180 end systems and other DetNet nodes. 182 DetNet flow 183 A DetNet flow is a sequence of packets to which the DetNet 184 service is to be provided. 186 DetNet compound flow and DetNet member flow 187 A DetNet compound flow is a DetNet flow that has been 188 separated into multiple duplicate DetNet member flows, which 189 are eventually merged back into a single DetNet compound 190 flow, at the DetNet transport layer. "Compound" and "member" 191 are strictly relative to each other, not absolutes; a DetNet 192 compound flow comprising multiple DetNet member flows can, in 193 turn, be a member of a higher-order compound. 195 DetNet intermediate node 196 A DetNet relay node or transit node. 198 DetNet edge node 199 An instance of a DetNet relay node that includes either a 200 DetNet service layer proxy function for DetNet service 201 protection (e.g. the addition or removal of packet sequencing 202 information) for one or more end systems, or starts or 203 terminates congestion protection at the DetNet transport 204 layer, analogous to a Label Edge Router (LER). 206 DetNet-UNI 207 User-to-Network Interface with DetNet specific 208 functionalities. It is a packet-based reference point and 209 may provide multiple functions like encapsulation, status, 210 synchronization, etc. 212 end system 213 Commonly called a "host" or "node" in IETF documents, and an 214 "end station" is IEEE 802 documents. End systems of interest 215 to this document are either sources or destinations of DetNet 216 flows. And end system may or may not be DetNet transport 217 layer aware or DetNet service layer aware. 219 link 220 A connection between two DetNet nodes. It may be composed of 221 a physical link or a sub-network technology that can provide 222 appropriate traffic delivery for DetNet flows. 224 DetNet node 225 A DetNet aware end system, transit node, or relay node. 226 "DetNet" may be omitted in some text. 228 Detnet relay node 229 A DetNet node including a service layer function that 230 interconnects different DetNet transport layer paths to 231 provide service protection. A DetNet relay node can be a 232 bridge, a router, a firewall, or any other system that 233 participates in the DetNet service layer. It typically 234 incorporates DetNet transport layer functions as well, in 235 which case it is collocated with a transit node. 237 reservation 238 A trail of configuration between source to destination(s) 239 through transit nodes and subnets associated with a DetNet 240 flow, to provide congestion protection. 242 DetNet service layer 243 The layer at which service protection is provided, either 244 packet sequencing, replication, and elimination 245 (Section 3.2.4) or network coding (Section 5.4). 247 source 248 An end system capable of sourcing a DetNet flow. 250 DetNet transit node 251 A node operating at the DetNet transport layer, that utilizes 252 link layer and/or network layer switching across multiple 253 links and/or sub-networks to provide paths for DetNet service 254 layer functions. Optionally provides congestion protection 255 over those paths. An MPLS LSR is an example of a DetNet 256 transit node. 258 DetNet transport layer 259 The layer that optionally provides congestion protection for 260 DetNet flows over paths provided by the underlying network. 262 TSN 263 Time-Sensitive Networking, TSN is a Task Group of the IEEE 264 802.1 Working Group. 266 2.2. IEEE 802 TSN to DetNet dictionary 268 This section also serves as a dictionary for translating from the 269 terms used by the IEEE 802 Time-Sensitive Networking (TSN) Task Group 270 to those of the DetNet WG. 272 Listener 273 The IEEE 802 term for a destination of a DetNet flow. 275 relay system 276 The IEEE 802 term for a DetNet intermediate node. 278 Stream 279 The IEEE 802 term for a DetNet flow. 281 Talker 282 The IEEE 802 term for the source of a DetNet flow. 284 3. Providing the DetNet Quality of Service 286 3.1. Primary goals defining the DetNet QoS 288 The DetNet Quality of Service can be expressed in terms of: 290 o Minimum and maximum end-to-end latency from source to destination; 291 timely delivery and jitter avoidance derive from these constraints 293 o Probability of loss of a packet, under various assumptions as to 294 the operational states of the nodes and links. A derived property 295 is whether it is acceptable to deliver a duplicate packet, which 296 is an inherent risk in highly reliable and/or broadcast 297 transmissions 299 It is a distinction of DetNet that it is concerned solely with worst- 300 case values for the end-to-end latency. Average, mean, or typical 301 values are of no interest, because they do not affect the ability of 302 a real-time system to perform its tasks. In general, a trivial 303 priority-based queuing scheme will give better average latency to a 304 data flow than DetNet, but of course, the worst-case latency can be 305 essentially unbounded. 307 Three techniques are used by DetNet to provide these qualities of 308 service: 310 o Congestion protection (Section 3.2.1). 312 o Explicit routes (Section 3.2.2). 314 o Service protection (Section 3.2.4). 316 Congestion protection operates by reserving resources along the path 317 of a DetNet Flow, e.g. buffer space or link bandwidth. Congestion 318 protection greatly reduces, or even eliminates entirely, packet loss 319 due to output packet congestion within the network, but it can only 320 be supplied to a DetNet flow that is limited at the source to a 321 maximum packet size and transmission rate. 323 Congestion protection addresses both of the DetNet QoS requirements 324 (latency and packet loss). Given that DetNet nodes have a finite 325 amount of buffer space, congestion protection necessarily results in 326 a maximum end-to-end latency. It also addresses the largest 327 contribution to packet loss, which is buffer congestion. 329 After congestion, the most important contributions to packet loss are 330 typically from random media errors and equipment failures. Service 331 protection is the name for the mechanisms used by DetNet to address 332 these losses. The mechanisms employed are constrained by the 333 requirement to meet the users' latency requirements. Packet 334 replication and elimination (Section 3.2.4) packet encoding 335 Section 5.4 are described in this document to provide service 336 protection; others may be found. Both mechanisms distribute the 337 contents of DetNet flows over multiple paths in time and/or space, so 338 that the loss of some of the paths does need not cause the loss of 339 any packets. The paths are typically (but not necessarily) explicit 340 routes, so that they cannot suffer temporary interruptions caused by 341 the reconvergence of routing or bridging protocols. 343 These three techniques can be applied independently, giving eight 344 possible combinations, including none (no DetNet), although some 345 combinations are of wider utility than others. This separation keeps 346 the protocol stack coherent and maximizes interoperability with 347 existing and developing standards in this (IETF) and other Standards 348 Development Organizations. Some examples of typical expected 349 combinations: 351 o Explicit routes plus service protection are exactly the techniques 352 employed by [HSR-PRP]. Explicit routes are achieved by limiting 353 the physical topology of the network, and the sequentialization, 354 replication, and duplicate elimination are facilitated by packet 355 tags added at the front or the end of Ethernet frames. 357 o Congestion protection alone is is offered by IEEE 802.1 Audio 358 Video bridging [IEEE802.1BA-2011]. As long as the network suffers 359 no failures, zero congestion loss can be achieved through the use 360 of a reservation protocol (MSRP), shapers in every bridge, and a 361 bit of network calculus. 363 o Using all three together gives maximum protection. 365 There are, of course, simpler methods available (and employed, today) 366 to achieve levels of latency and packet loss that are satisfactory 367 for many applications. Prioritization and over-provisioning is one 368 such technique. However, these methods generally work best in the 369 absence of any significant amount of non-critical traffic in the 370 network (if, indeed, such traffic is supported at all), or work only 371 if the critical traffic constitutes only a small portion of the 372 network's theoretical capacity, or work only if all systems are 373 functioning properly, or in the absence of actions by end systems 374 that disrupt the network's operations. 376 There are any number of methods in use, defined, or in progress for 377 accomplishing each of the above techniques. It is expected that this 378 DetNet Architecture will assist various vendors, users, and/or 379 "vertical" Standards Development Organizations (dedicated to a single 380 industry) to make selections among the available means of 381 implementing DetNet networks. 383 3.2. Mechanisms to achieve DetNet Qos 385 3.2.1. Congestion protection 387 The primary means by which DetNet achieves its QoS assurances is to 388 reduce, or even completely eliminate, congestion at an output port as 389 a cause of packet loss. Given that a DetNet flow cannot be 390 throttled, this can be achieved only by the provision of sufficient 391 buffer storage at each hop through the network to ensure that no 392 packets are dropped due to a lack of buffer storage. 394 Ensuring adequate buffering requires, in turn, that the source, and 395 every intermediate node along the path to the destination (or nearly 396 every node -- see Section 4.3.3) be careful to regulate its output to 397 not exceed the data rate for any DetNet flow, except for brief 398 periods when making up for interfering traffic. Any packet sent 399 ahead of its time potentially adds to the number of buffers required 400 by the next hop, and may thus exceed the resources allocated for a 401 particular DetNet flow. 403 The low-level mechanisms described in Section 4.5 provide the 404 necessary regulation of transmissions by an end system or 405 intermediate node to provide congestion protection. The reservation 406 of the bandwidth and buffers for a DetNet flow requires the 407 provisioning described in Section 4.9. A DetNet node may have other 408 resources requiring allocation and/or scheduling, that might 409 otherwise be over-subscribed and trigger the rejection of a 410 reservation. 412 3.2.2. Explicit routes 414 In networks controlled by typical peer-to-peer protocols such as IEEE 415 802.1 ISIS bridged networks or IETF OSPF routed networks, a network 416 topology event in one part of the network can impact, at least 417 briefly, the delivery of data in parts of the network remote from the 418 failure or recovery event. Thus, even redundant paths through a 419 network, if controlled by the typical peer-to-peer protocols, do not 420 eliminate the chances of brief losses of contact. 422 Many real-time networks rely on physical rings or chains of two-port 423 devices, with a relatively simple ring control protocol. This 424 supports redundant paths for service protection with a minimum of 425 wiring. As an additional benefit, ring topologies can often utilize 426 different topology management protocols than those used for a mesh 427 network, with a consequent reduction in the response time to topology 428 changes. Of course, this comes at some cost in terms of increased 429 hop count, and thus latency, for the typical path. 431 In order to get the advantages of low hop count and still ensure 432 against even very brief losses of connectivity, DetNet employs 433 explicit routes, where the path taken by a given DetNet flow does not 434 change, at least immediately, and likely not at all, in response to 435 network topology events. Service protection (Section 3.2.4 or 436 Section 5.4) over explicit routes provides a high likelihood of 437 continuous connectivity. Explicit routes are commonly used in MPLS 438 TE LSPs. 440 3.2.3. Jitter Reduction 442 A core objective of DetNet is to enable the convergence of Non-IP 443 networks onto a common network infrastructure. This requires the 444 accurate emulation of currently deployed mission-specific networks, 445 which typically rely on point-to-point analog (e.g. 4-20mA 446 modulation) and serial-digital cables (or buses) for highly reliable, 447 synchronized and jitter-free communications. While the latency of 448 analog transmissions is basically the speed of light, legacy serial 449 links are usually slow (in the order of Kbps) compared to, say, GigE, 450 and some latency is usually acceptable. What is not acceptable is 451 the introduction of excessive jitter, which may, for instance, affect 452 the stability of control systems. 454 Applications that are designed to operate on serial links usually do 455 not provide services to recover the jitter, because jitter simply 456 does not exists there. Streams of information are expected to be 457 delivered in-order and the precise time of reception influences the 458 processes. In order to converge such existing applications, there is 459 a desire to emulate all properties of the serial cable, such as clock 460 transportation, perfect flow isolation and fixed latency. While 461 minimal jitter (in the form of specifying minimum, as well as 462 maximum, end-to-end latency) is supported by DetNet, there are 463 practical limitations on packet-based networks in this regard. In 464 general, users are encouraged to use, instead of, "do this when you 465 get the packet," a combination of: 467 o Sub-microsecond time synchronization among all source and 468 destination end systems, and 470 o Time-of-execution fields in the application packets. 472 Jitter reduction is provided by the mechanisms described in 473 Section 4.5 that also provide congestion protection. 475 3.2.4. Packet Replication and Elimination 477 After congestion loss has been eliminated, the most important causes 478 of packet loss are random media and/or memory faults, and equipment 479 failures. Both causes of packet loss can be greatly reduced by 480 spreading the data in a packet over multiple transmissions. One such 481 method for service protection is described in this section, which 482 sends the same packets over multiple paths. See also Section 5.4. 484 Packet replication and elimination, also known as seamless redundancy 485 [HSR-PRP], or 1+1 hitless protection, is a function of the DetNet 486 service layer. It involves three capabilities: 488 o Providing sequencing information, once, at or near the source, to 489 the packets of a DetNet compound flow. This may be done by adding 490 a sequence number or time stamp as part of DetNet, or may be 491 inherent in the packet, e.g. in a transport protocol, or 492 associated to other physical properties such as the precise time 493 (and radio channel) of reception of the packet. Section 3.2.2. 495 o Replicating these packets into multiple DetNet member flows and, 496 typically, sending them along at least two different paths to the 497 destination(s), e.g. over the explicit routes of 499 o Eliminating duplicated packets. This may be done at any step 500 along the path to save network resources further down, in 501 particular if multiple Replication points exist. But the most 502 common case is to perform this operation at the very edge of the 503 DetNet network, preferably in or near the receiver. 505 This function is a "hitless" version of, e.g., the 1+1 linear 506 protection in [RFC6372]. That is, instead of switching from one flow 507 to the other when a failure of a flow is detected, DetNet combines 508 both flows, and performs a packet-by-packet selection of which to 509 discard, based on sequence number. 511 In the simplest case, this amounts to replicating each packet in a 512 source that has two interfaces, and conveying them through the 513 network, along separate paths, to the similarly dual-homed 514 destinations, that discard the extras. This ensures that one path 515 (with zero congestion loss) remains, even if some intermediate node 516 fails. The sequence numbers can also be used for loss detection and 517 for re-ordering. 519 Detnet relay nodes in the network can provide replication and 520 elimination facilities at various points in the network, so that 521 multiple failures can be accommodated. 523 This is shown in the following figure, where the two relay nodes each 524 replicate (R) the DetNet flow on input, sending the DetNet member 525 flows to both the other relay node and to the end system, and 526 eliminate duplicates (E) on the output interface to the right-hand 527 end system. Any one link in the network can fail, and the Detnet 528 compound flow can still get through. Furthermore, two links can 529 fail, as long as they are in different segments of the network. 531 Packet replication and elimination 533 > > > > > > > > > relay > > > > > > > > 534 > /------------+ R node E +------------\ > 535 > / v + ^ \ > 536 end R + v | ^ + E end 537 system + v | ^ + system 538 > \ v + ^ / > 539 > \------------+ R relay E +-----------/ > 540 > > > > > > > > > node > > > > > > > > 542 Figure 1 544 Note that packet replication and elimination does not react to and 545 correct failures; it is entirely passive. Thus, intermittent 546 failures, mistakenly created packet filters, or misrouted data is 547 handled just the same as the equipment failures that are detected 548 handled by typical routing and bridging protocols. 550 If packet replication and elimination is used over paths providing 551 congestion protection (Section 3.2.1), and member flows that take 552 different-length paths through the network are combined, a merge 553 point may require extra buffering to equalize the delays over the 554 different paths. This equalization ensures that the resultant 555 compound flow will not exceed its contracted bandwidth even after one 556 or the other of the paths is restored after a failure. 558 3.3. Secondary goals for DetNet 560 Many applications require DetNet to provide additional services, 561 including coesistence with other QoS mechanisms Section 3.3.1 and 562 protection against misbehaving transmitters Section 3.3.2. 564 3.3.1. Coexistence with normal traffic 566 A DetNet network supports the dedication of a high proportion (e.g. 567 75%) of the network bandwidth to DetNet flows. But, no matter how 568 much is dedicated for DetNet flows, it is a goal of DetNet to coexist 569 with existing Class of Service schemes (e.g., DiffServ). It is also 570 important that non-DetNet traffic not disrupt the DetNet flow, of 571 course (see Section 3.3.2 and Section 6). For these reasons: 573 o Bandwidth (transmission opportunities) not utilized by a DetNet 574 flow are available to non-DetNet packets (though not to other 575 DetNet flows). 577 o DetNet flows can be shaped or scheduled, in order to ensure that 578 the highest-priority non-DetNet packet also is ensured a worst- 579 case latency (at any given hop). 581 o When transmission opportunities for DetNet flows are scheduled in 582 detail, then the algorithm constructing the schedule should leave 583 sufficient opportunities for non-DetNet packets to satisfy the 584 needs of the users of the network. Detailed scheduling can also 585 permit the time-shared use of buffer resources by different DetNet 586 flows. 588 Ideally, the net effect of the presence of DetNet flows in a network 589 on the non-DetNet packets is primarily a reduction in the available 590 bandwidth. 592 3.3.2. Fault Mitigation 594 One key to building robust real-time systems is to reduce the 595 infinite variety of possible failures to a number that can be 596 analyzed with reasonable confidence. DetNet aids in the process by 597 providing filters and policers to detect DetNet packets received on 598 the wrong interface, or at the wrong time, or in too great a volume, 599 and to then take actions such as discarding the offending packet, 600 shutting down the offending DetNet flow, or shutting down the 601 offending interface. 603 It is also essential that filters and service remarking be employed 604 at the network edge to prevent non-DetNet packets from being mistaken 605 for DetNet packets, and thus impinging on the resources allocated to 606 DetNet packets. 608 There exist techniques, at present and/or in various stages of 609 standardization, that can perform these fault mitigation tasks that 610 deliver a high probability that misbehaving systems will have zero 611 impact on well-behaved DetNet flows, except of course, for the 612 receiving interface(s) immediately downstream of the misbehaving 613 device. Examples of such techniques include traffic policing 614 functions (e.g. [RFC2475]) and separating flows into per-flow rate- 615 limited queues. 617 4. DetNet Architecture 619 4.1. DetNet stack model 621 4.1.1. Representative Protocol Stack Model 623 Figure 2 illustrates a conceptual DetNet data plane layering model. 624 One may compare it to that in [IEEE802.1CB], Annex C, a work in 625 progress. 627 DetNet data plane protocol stack 629 | packets going | ^ packets coming ^ 630 v down the stack v | up the stack | 631 +----------------------+ +-----------------------+ 632 | Source | | Destination | 633 +----------------------+ +-----------------------+ 634 | Service layer | | Service layer | 635 | Packet sequencing | | Duplicate elimination | 636 | Flow duplication | | Flow merging | 637 | Packet encoding | | Packet decoding | 638 +----------------------+ +-----------------------+ 639 | Transport layer | | Transport layer | 640 | Congestion prot. | | Congestion prot. | 641 +----------------------+ +-----------------------+ 642 | Lower layers | | Lower layers | 643 +----------------------+ +-----------------------+ 644 v ^ 645 \_________________________/ 647 Figure 2 649 Not all layers are required for any given application, or even for 650 any given network. The layers are, from top to bottom: 652 Application 653 Shown as "source" and "destination" in the diagram. 655 OAM 656 Operations, Administration, and Maintenance leverages in-band 657 and out-of-and signaling that validates whether the service 658 is effectively obtained within QoS constraints. OAM is not 659 shown in Figure 2; it may reside in any number of the layers. 660 OAM can involve specific tagging added in the packets for 661 tracing implementation or network configuration errors; 662 traceability enables to find whether a packet is a replica, 663 which relay node performed the replication, and which segment 664 was intended for the replica. 666 Packet sequencing 667 As part of DetNet service protection, supplies the sequence 668 number for packet replication and elimination 669 (Section 3.2.4). Peers with Duplicate elimination. This 670 layer is not needed if a higher-layer transport protocol is 671 expected to perform any packet sequencing and duplicate 672 elimination required by the DetNet flow duplication. 674 Duplicate elimination 675 As part of the DetNet service layer, based on the sequenced 676 number supplied by its peer, packet sequencing, Duplicate 677 elimination discards any duplicate packets generated by 678 DetNet flow duplication. It can operate on member flows, 679 compound flows, or both. The duplication may also be 680 inferred from other information such as the precise time of 681 reception in a scheduled network. The duplicate elimination 682 layer may also perform resequencing of packets to restore 683 packet order in a flow that was disrupted by the loss of 684 packets on one or another of the multiple paths taken. 686 Flow duplication 687 As part of DetNet service protection, replicates packets that 688 belong to a DetNet compound flow into two or more DetNet 689 member flows. Note that this function is separate from 690 packet sequencing. Flow duplication can be an explicit 691 duplication and remarking of packets, or can be performed by, 692 for example, techniques similar to ordinary multicast 693 replication. Peers with DetNet flow merging. 695 Network flow merging 696 As part of DetNet service protection, merges DetNet member 697 flows together for packets coming up the stack belonging to a 698 specific DetNet compound flow. Peers with DetNet flow 699 duplication. DetNet flow merging, together with packet 700 sequencing, duplicate elimination, and DetNet flow 701 duplication, performs packet replication and elimination 702 (Section 3.2.4). 704 Packet encoding 705 As part of DetNet service protection, as an alternative to 706 packet sequencing and flow duplication, packet encoding 707 combines the information in multiple DetNet packets, perhaps 708 from different DetNet compound flows, and transmits that 709 information in packets on different DetNet member Flows. 710 Peers with Packet decoding. 712 Packet decoding 713 As part of DetNet service protection, as an alternative to 714 flow merging and duplicate elimination, packet decoding takes 715 packets from different DetNet member flows, and computes from 716 those packets the original DetNet packets from the compound 717 flows input to packet encoding. Peers with Packet encoding. 719 Congestion protection 720 The DetNet transport layer provides congestion protection. 721 See Section 4.5. The actual queuing and shaping mechanisms 722 are typically provided by underlying subnet layers, but since 723 these are can be closely associated with the means of 724 providing paths for DetNet flows (e.g. MPLS LSPs or {VLAN, 725 multicast destination MAC address} pairs), the path and the 726 congestion protection are conflated in this figure. 728 Note that the packet sequencing and duplication elimination functions 729 at the source and destination ends of a DetNet compound flow may be 730 performed either in the end system or in a DetNet edge node. The 731 reader must not confuse a DetNet edge function with other kinds of 732 edge functions, e.g. an Label Edge Router, although the two functions 733 may be performed together. The DetNet edge function is concerned 734 with sequencing packets belonging to DetNet flows. The LER with 735 encapsulating/decapsulating packets for transport, and is considered 736 part of the network underlying the DetNet transport layer. 738 4.1.2. DetNet Data Plane Overview 740 A "Deterministic Network" will be composed of DetNet enabled nodes 741 i.e., End Systems, Edge Nodes, Relay Nodes and collectively deliver 742 DetNet services. DetNet enabled nodes are interconnected via Transit 743 Nodes (i.e., routers) which support DetNet, but are not DetNet 744 service aware. Transit nodes see DetNet nodes as end points. All 745 DetNet enabled nodes are connect to sub-networks, where a point-to- 746 point link is also considered as a simple sub-network. These sub- 747 networks will provide DetNet compatible service for support of DetNet 748 traffic. Examples of sub-networks include IEEE 802.1 TSN and OTN. 749 Of course, multi-layer DetNet systems may also be possible, where one 750 DetNet appears as a sub-network, and provides service to, a higher 751 layer DetNet system. A simple DetNet concept network is shown in 752 Figure 3. 754 TSN Edge Transit Relay DetNet 755 End System Node Node Node End System 757 +---------+ +.........+ +---------+ 758 | Appl. |<---:Svc Proxy:-- End to End Service ---------->| Appl. | 759 +---------+ +---------+ +---------+ +---------+ 760 | TSN | |TSN| |Svc|<-- DetNet flow ---: Service :-->| Service | 761 +---------+ +---+ +---+ +---------+ +---------+ +---------+ 762 |Transport| |Trp| |Trp| |Transport| |Trp| |Trp| |Transport| 763 +-------.-+ +-.-+ +-.-+ +--.----.-+ +-.-+ +-.-+ +---.-----+ 764 : Link : / ,-----. \ : Link : / ,-----. \ 765 +........+ +-[ Sub ]-+ +........+ +-[ Sub ]-+ 766 [Network] [Network] 767 `-----' `-----' 769 Figure 3: A Simple DetNet Enabled Network 771 Distinguishing the function of these two DetNet data plane layers, 772 the DetNet service layer and the DetNet transport layer, helps to 773 explore and evaluate various combinations of the data plane solutions 774 available. This separation of DetNet layers, while helpful, should 775 not be considered as formal requirement. For example, some 776 technologies may violate these strict layers and still be able to 777 deliver a DetNet service. 779 . 780 . 781 +-----------+ 782 | Service | PW, RTP/(UDP), GRE 783 +-----------+ 784 | Transport | (UDP)/IPv6, (UDP)/IPv4, MPLS LSPs, BIER 785 +-----------+ 786 . 787 . 789 Figure 4: DetNet adaptation to data plane 791 In some networking scenarios, the end system initially provides a 792 DetNet flow encapsulation, which contains all information needed by 793 DetNet nodes (e.g., Real-time Transport Protocol (RTP) [RFC3550] 794 based DetNet flow transported over a native UDP/IP network or 795 PseudoWire). In other scenarios, the encapsulation formats might 796 differ significantly. As an example, a CPRI "application's" I/Q data 797 mapped directly to Ethernet frames may have to be transported over an 798 MPLS-based packet switched network (PSN). 800 There are many valid options to create a data plane solution for 801 DetNet traffic by selecting a technology approach for the DetNet 802 service layer and also selecting a technology approach for the DetNet 803 transport layer. There are a high number of valid combinations. 805 One of the most fundamental differences between different potential 806 data plane options is the basic addressing and headers used by DetNet 807 end systems. For example, is the basic service a Layer 2 (e.g., 808 Ethernet) or Layer 3 (i.e., IP) service. This decision impacts how 809 DetNet end systems are addressed, and the basic forwarding logic for 810 the DetNet service layer. 812 4.1.3. Network reference model 814 The figure below shows another view of the DetNet service related 815 reference points and main components (Figure 5). 817 DetNet DetNet 818 end system end system 819 _ _ 820 / \ +----DetNet-UNI (U) / \ 821 /App\ | /App\ 822 /-----\ | /-----\ 823 | NIC | v ________ | NIC | 824 +--+--+ _____ / \ DetNet-UNI (U) --+ +--+--+ 825 | / \__/ \ | | 826 | / +----+ +----+ \_____ | | 827 | / | | | | \_______ | | 828 +------U PE +----+ P +----+ \ _ v | 829 | | | | | | | ___/ \ | 830 | +--+-+ +----+ | +----+ | / \_ | 831 \ | | | | | / \ | 832 \ | +----+ +--+-+ +--+PE |-------- U------+ 833 \ | | | | | | | | | \_ _/ 834 \ +---+ P +----+ P +--+ +----+ | \____/ 835 \___ | | | | / 836 \ +----+__ +----+ DetNet-1 DetNet-2 837 | \_____/ \___________/ | 838 | | 839 | | End-to-End-Service | | | | 840 <----------------------------------------------------------------> 841 | | DetNet-Service | | | | 842 | <--------------------------------------------------> | 843 | | | | | | 845 Figure 5: DetNet Service Reference Model (multi-domain) 847 DetNet-UNIs ("U" in Figure 5) are assumed in this document to be 848 packet-based reference points and provide connectivity over the 849 packet network. A DetNet-UNI may provide multiple functions, e.g., 850 it may add networking technology specific encapsulation to the DetNet 851 flows if necessary; it may provide status of the availability of the 852 connection associated to a reservation; it may provide a 853 synchronization service for the end system; it may carry enough 854 signaling to place the reservation in a network without a controller, 855 or if the controller only deals with the network but not the end 856 points. Internal reference points of end systems (between the 857 application and the NIC) are more challenging from control 858 perspective and they may have extra requirements (e.g., in-order 859 delivery is expected in end system internal reference points, whereas 860 it is considered optional over the DetNet-UNI), therefore not covered 861 in this document. 863 4.2. DetNet systems 865 4.2.1. End system 867 The native data flow between the source/destination end systems is 868 referred to as application-flow (App-flow). The traffic 869 characteristics of an App-flow can be CBR (constant bit rate) or VBR 870 (variable bit rate) and can have L1 or L2 or L3 encapsulation (e.g., 871 TDM (time-division multiplexing), Ethernet, IP). These 872 characteristics are considered as input for resource reservation and 873 might be simplified to ensure determinism during transport (e.g., 874 making reservations for the peak rate of VBR traffic, etc.). 876 An end system may or may not be DetNet transport layer aware or 877 DetNet service layer aware. That is, an end system may or may not 878 contain DetNet specific functionality. End systems with DetNet 879 functionalities may have the same or different transport layer as the 880 connected DetNet domain. Grouping of end systems are shown in 881 Figure 6. 883 End system 884 | 885 | 886 | DetNet aware ? 887 / \ 888 +------< >------+ 889 NO | \ / | YES 890 | v | 891 DetNet unaware | 892 End system | 893 | Service/ 894 | Transport 895 / \ aware ? 896 +--------< >-------------+ 897 t-aware | \ / | s-aware 898 | v | 899 | | both | 900 | | | 901 DetNet t-aware | DetNet s-aware 902 End system | End system 903 v 904 DetNet st-aware 905 End system 907 Figure 6: Grouping of end systems 909 Note some known use cases for end systems: 911 o DetNet unaware: The classic case requiring network proxies. 913 o DetNet t-aware: An extant TSN system. It knows about some TSN 914 functions (e.g., reservation), but not about replication/ 915 elimination. 917 o DetNet s-aware: An extant IEC 62439-3 system. It supplies 918 sequence numbers, but doesn't know about zero congestion loss. 920 o DetNet st-aware: A full functioning DetNet end station, it has 921 DetNet functionalities and usually the same forwarding paradigm as 922 the connected DetNet domain. It can be treated as an integral 923 part of the DetNet domain . 925 4.2.2. DetNet edge, relay, and transit nodes 927 As shown in Figure 3, DetNet edge nodes providing proxy service and 928 DetNet relay nodes providing the DetNet service layer are DetNet- 929 aware, and DetNet transit nodes need only be aware of the DetNet 930 transport layer. 932 In general, if a DetNet flow passes through one or more DetNet- 933 unaware network node between two DetNet nodes providing the DetNet 934 transport layer for that flow, there is a potential for disruption or 935 failure of the DetNet QoS. A network administrator needs to ensure 936 that the DetNet-unaware network nodes are configured to minimize the 937 chances of packet loss and delay, and provision enough exra buffer 938 space in the DetNet transit node following the DetNet-unaware network 939 nodes to absorb the induced latency variations. 941 4.3. DetNet flows 943 4.3.1. DetNet flow types 945 A DetNet flow can have different formats during while it is 946 transported between the peer end systems. Therefore, the following 947 possible types / formats of a DetNet flow are distinguished in this 948 document: 950 o App-flow: native format of a DetNet flow. It does not contain any 951 DetNet related attributes. 953 o DetNet-t-flow: specific format of a DetNet flow. Only requires 954 the congestion / latency features provided by the Detnet transport 955 layer. 957 o DetNet-s-flow: specific format of a DetNet flow. Only requires 958 the replication/elimination feature ensured by the DetNet service 959 layer. 961 o DetNet-st-flow: specific format of a DetNet flow. It requires 962 both DetNet service layer and DetNet transport layer functions 963 during forwarding. 965 4.3.2. Source guarantees 967 For the purposes of congestion protection, DetNet flows can be 968 synchronous or asynchronous. In synchronous DetNet flows, at least 969 the intermediate nodes (and possibly the end systems) are closely 970 time synchronized, typically to better than 1 microsecond. By 971 transmitting packets from different DetNet flows or classes of DetNet 972 flows at different times, using repeating schedules synchronized 973 among the intermediate nodes, resources such as buffers and link 974 bandwidth can be shared over the time domain among different DetNet 975 flows. There is a tradeoff among techniques for synchronous DetNet 976 flows between the burden of fine-grained scheduling and the benefit 977 of reducing the required resources, especially buffer space. 979 In contrast, asynchronous DetNet flows are not coordinated with a 980 fine-grained schedule, so relay and end systems must assume worst- 981 case interference among DetNet flows contending for buffer resources. 982 Asynchronous DetNet flows are characterized by: 984 o A maximum packet size; 986 o An observation interval; and 988 o A maximum number of transmissions during that observation 989 interval. 991 These parameters, together with knowledge of the protocol stack used 992 (and thus the size of the various headers added to a packet), limit 993 the number of bit times per observation interval that the DetNet flow 994 can occupy the physical medium. 996 The source promises that these limits will not be exceeded. If the 997 source transmits less data than this limit allows, the unused 998 resources such as link bandwidth can be made available by the system 999 to non-DetNet packets. However, making those resources available to 1000 DetNet packets in other DetNet flows would serve no purpose. Those 1001 other DetNet flows have their own dedicated resources, on the 1002 assumption that all DetNet flows can use all of their resources over 1003 a long period of time. 1005 Note that there is no provision in DetNet for throttling DetNet flows 1006 (reducing the transmission rate via feedback); the assumption is that 1007 a DetNet flow, to be useful, must be delivered in its entirety. That 1008 is, while any useful application is written to expect a certain 1009 number of lost packets, the real-time applications of interest to 1010 DetNet demand that the loss of data due to the network is 1011 extraordinarily infrequent. 1013 Although DetNet strives to minimize the changes required of an 1014 application to allow it to shift from a special-purpose digital 1015 network to an Internet Protocol network, one fundamental shift in the 1016 behavior of network applications is impossible to avoid: the 1017 reservation of resources before the application starts. In the first 1018 place, a network cannot deliver finite latency and practically zero 1019 packet loss to an arbitrarily high offered load. Secondly, achieving 1020 practically zero packet loss for unthrottled (though bandwidth 1021 limited) DetNet flows means that bridges and routers have to dedicate 1022 buffer resources to specific DetNet flows or to classes of DetNet 1023 flows. The requirements of each reservation have to be translated 1024 into the parameters that control each system's queuing, shaping, and 1025 scheduling functions and delivered to the hosts, bridges, and 1026 routers. 1028 4.3.3. Incomplete Networks 1030 The presence in the network of transit nodes or subnets that are not 1031 fully capable of offering DetNet services complicates the ability of 1032 the intermediate nodes and/or controller to allocate resources, as 1033 extra buffering, and thus extra latency, must be allocated at points 1034 downstream from the non-DetNet intermediate node for a DetNet flow. 1036 4.4. Traffic Engineering for DetNet 1038 Traffic Engineering Architecture and Signaling (TEAS) [TEAS] defines 1039 traffic-engineering architectures for generic applicability across 1040 packet and non-packet networks. From TEAS perspective, Traffic 1041 Engineering (TE) refers to techniques that enable operators to 1042 control how specific traffic flows are treated within their networks. 1044 Because if its very nature of establishing explicit optimized paths, 1045 Deterministic Networking can be seen as a new, specialized branch of 1046 Traffic Engineering, and inherits its architecture with a separation 1047 into planes. 1049 The Deterministic Networking architecture is thus composed of three 1050 planes, a (User) Application Plane, a Controller Plane, and a Network 1051 Plane, which echoes that of Figure 1 of Software-Defined Networking 1052 (SDN): Layers and Architecture Terminology [RFC7426].: 1054 4.4.1. The Application Plane 1056 Per [RFC7426], the Application Plane includes both applications and 1057 services. In particular, the Application Plane incorporates the User 1058 Agent, a specialized application that interacts with the end user / 1059 operator and performs requests for Deterministic Networking services 1060 via an abstract Flow Management Entity, (FME) which may or may not be 1061 collocated with (one of) the end systems. 1063 At the Application Plane, a management interface enables the 1064 negotiation of flows between end systems. An abstraction of the flow 1065 called a Traffic Specification (TSpec) provides the representation. 1066 This abstraction is used to place a reservation over the (Northbound) 1067 Service Interface and within the Application plane. It is associated 1068 with an abstraction of location, such as IP addresses and DNS names, 1069 to identify the end systems and eventually specify intermediate 1070 nodes. 1072 4.4.2. The Controller Plane 1074 The Controller Plane corresponds to the aggregation of the Control 1075 and Management Planes in [RFC7426], though Common Control and 1076 Measurement Plane (CCAMP) [CCAMP] makes an additional distinction 1077 between management and measurement. When the logical separation of 1078 the Control, Measurement and other Management entities is not 1079 relevant, the term Controller Plane is used for simplicity to 1080 represent them all, and the term controller refers to any device 1081 operating in that plane, whether is it a Path Computation entity or a 1082 Network Management entity (NME). The Path Computation Element (PCE) 1083 [PCE] is a core element of a controller, in charge of computing 1084 Deterministic paths to be applied in the Network Plane. 1086 A (Northbound) Service Interface enables applications in the 1087 Application Plane to communicate with the entities in the Controller 1088 Plane. 1090 One or more PCE(s) collaborate to implement the requests from the FME 1091 as Per-Flow Per-Hop Behaviors installed in the intermediate nodes for 1092 each individual flow. The PCEs place each flow along a deterministic 1093 sequence of intermediate nodes so as to respect per-flow constraints 1094 such as security and latency, and optimize the overall result for 1095 metrics such as an abstract aggregated cost. The deterministic 1096 sequence can typically be more complex than a direct sequence and 1097 include redundancy path, with one or more packet replication and 1098 elimination points. 1100 4.4.3. The Network Plane 1102 The Network Plane represents the network devices and protocols as a 1103 whole, regardless of the Layer at which the network devices operate. 1104 It includes Forwarding Plane (data plane), Application, and 1105 Operational Plane (control plane) aspects. 1107 The network Plane comprises the Network Interface Cards (NIC) in the 1108 end systems, which are typically IP hosts, and intermediate nodes, 1109 which are typically IP routers and switches. Network-to-Network 1110 Interfaces such as used for Traffic Engineering path reservation in 1111 [RFC5921], as well as User-to-Network Interfaces (UNI) such as 1112 provided by the Local Management Interface (LMI) between network and 1113 end systems, are both part of the Network Plane, both in the control 1114 plane and the data plane. 1116 A Southbound (Network) Interface enables the entities in the 1117 Controller Plane to communicate with devices in the Network Plane. 1118 This interface leverages and extends TEAS to describe the physical 1119 topology and resources in the Network Plane. 1121 Flow Management Entity 1123 End End 1124 System System 1126 -+-+-+-+-+-+-+ Northbound -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1128 PCE PCE PCE PCE 1130 -+-+-+-+-+-+-+ Southbound -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1132 intermediate intermed. intermed. intermed. 1133 Node Node Node Node 1134 NIC NIC 1135 intermediate intermed. intermed. intermed. 1136 Node Node Node Node 1138 Figure 7 1140 The intermediate nodes (and eventually the end systems NIC) expose 1141 their capabilities and physical resources to the controller (the 1142 PCE), and update the PCE with their dynamic perception of the 1143 topology, across the Southbound Interface. In return, the PCE(s) set 1144 the per-flow paths up, providing a Flow Characterization that is more 1145 tightly coupled to the intermediate node Operation than a TSpec. 1147 At the Network plane, intermediate nodes may exchange information 1148 regarding the state of the paths, between adjacent systems and 1149 eventually with the end systems, and forward packets within 1150 constraints associated to each flow, or, when unable to do so, 1151 perform a last resort operation such as drop or declassify. 1153 This specification focuses on the Southbound interface and the 1154 operation of the Network Plane. 1156 4.5. Queuing, Shaping, Scheduling, and Preemption 1158 DetNet achieves congestion protection and bounded delivery latency by 1159 reserving bandwidth and buffer resources at every hop along the path 1160 of the DetNet flow. The reservation itself is not sufficient, 1161 however. Implementors and users of a number of proprietary and 1162 standard real-time networks have found that standards for specific 1163 data plane techniques are required to enable these assurances to be 1164 made in a multi-vendor network. The fundamental reason is that 1165 latency variation in one system results in the need for extra buffer 1166 space in the next-hop system(s), which in turn, increases the worst- 1167 case per-hop latency. 1169 Standard queuing and transmission selection algorithms allow a 1170 central controller to compute the latency contribution of each 1171 transit node to the end-to-end latency, to compute the amount of 1172 buffer space required in each transit node for each incremental 1173 DetNet flow, and most importantly, to translate from a flow 1174 specification to a set of values for the managed objects that control 1175 each relay or end system. The IEEE 802 has specified (and is 1176 specifying) a set of queuing, shaping, and scheduling algorithms that 1177 enable each transit node (bridge or router), and/or a central 1178 controller, to compute these values. These algorithms include: 1180 o A credit-based shaper [IEEE802.1Q-2014] Clause 34. 1182 o Time-gated queues governed by a rotating time schedule, 1183 synchronized among all transit nodes [IEEE802.1Qbv]. 1185 o Synchronized double (or triple) buffers driven by synchronized 1186 time ticks. [IEEE802.1Qch]. 1188 o Pre-emption of an Ethernet packet in transmission by a packet with 1189 a more stringent latency requirement, followed by the resumption 1190 of the preempted packet [IEEE802.1Qbu], [IEEE802.3br]. 1192 While these techniques are currently embedded in Ethernet and 1193 bridging standards, we can note that they are all, except perhaps for 1194 packet preemption, equally applicable to other media than Ethernet, 1195 and to routers as well as bridges. 1197 4.6. Service instance 1199 [Note: Service instance represents all the functions required on a 1200 node to allow the end-to-end service between the UNIs.] 1202 The DetNet network reference model is shown in Figure 8 for a DetNet- 1203 Service scenario (i.e. between two DetNet-UNIs). In this figure, the 1204 end systems ("A" and "B") are connected directly to the edge nodes of 1205 the IP/MPLS network ("PE1" and "PE2"). End-systems participating 1206 DetNet communication may require connectivity before setting up an 1207 App-flow that requires the DetNet service. Such a connectivity 1208 related service instance and the one dedicated for DetNet service 1209 share the same access. Packets belonging to a DetNet flow are 1210 selected by a filter configured on the access ("F1" and "F2"). As a 1211 result, data flow specific access ("access-A + F1" and "access-B + 1212 F2") are terminated in the flow specific service instance ("SI-1" and 1213 "SI-2"). A tunnel is used to provide connectivity between the 1214 service instances. 1216 The tunnel is used to transport exclusively the packets of the DetNet 1217 flow between "SI-1" and "SI-2". The service instances are configured 1218 to implement DetNet functions and a flow specific routing or bridging 1219 function depending on what connectivity the participating end systems 1220 require (L3 or L2). The service instance and the tunnel may or may 1221 not be shared by multiple DetNet flows. Sharing the service instance 1222 by multiple DetNet flows requires properly populated forwarding 1223 tables of the service instance. 1225 access-A access-B 1226 <-----> <---------- tunnel ----------> <-----> 1228 +---------+ ___ _ +---------+ 1229 End system | +----+ | / \/ \_ | +----+ | End system 1230 "A" -------F1+ | | / \ | | +F2----- "B" 1231 | | +==========+ IP/MPLS +========+ | | 1232 | |SI-1| | \__ Net._/ | |SI-2| | 1233 | +----+ | \____/ | +----+ | 1234 |PE1 | | PE2| 1235 +---------+ +---------+ 1237 Figure 8: DetNet network reference model 1239 [Note: The tunnel between the service instances may have some special 1240 characteristics. For example, in case of a "packet PW" based tunnel, 1241 there are differences in the usage of the packet PW for DetNet 1242 traffic compared to the network model described in [RFC6658]. In the 1243 DetNet scenario, the packet PW is used exclusively by the DetNet 1244 flow, whereas [RFC6658] states: "The packet PW appears as a single 1245 point-to-point link to the client layer. Network-layer adjacency 1246 formation and maintenance between the client equipments will follow 1247 the normal practice needed to support the required relationship in 1248 the client layer ... This packet pseudowire is used to transport all 1249 of the required layer 2 and layer 3 protocols between LSR1 and 1250 LSR2".] 1252 [Note: Examples are provided in Annex 1 of 1253 [I-D.varga-detnet-service-model].] 1255 4.7. Flow identification at technology borders 1257 4.7.1. Exporting flow identification 1259 An interesting feature of DetNet, and one that invites 1260 implementations that can be accused of "layering violations", is the 1261 need for lower layers to be aware of specific flows at higher layers, 1262 in order to provide specific queuing and shaping services for 1263 specific flows. For example: 1265 o A non-IP, strictly L2 source end system X may be sending multiple 1266 flows to the same L2 destination end system Y. Those flows may 1267 include DetNet flows with different QoS requirements, and may 1268 include non-DetNet flows. 1270 o A router may be sending any number of flows to another router. 1271 Again, those flows may include DetNet flows with different QoS 1272 requirements, and may include non-DetNet flows. 1274 o Two routers may be separated by bridges. For these bridges to 1275 perform any required per-flow queuing and shaping, they must be 1276 able to identify the individual flows. 1278 o A Label Edge Router (LERs) may have a Label Switched Path (LSP) 1279 set up for handling traffic destined for a particular IP address 1280 carrying only non-DetNet flows. If a DetNet flow to that same 1281 address is requested, a separate LSP may be needed, in order that 1282 all of the Label Switch Routers (LSRs) along the path to the 1283 destination give that flow special queuing and shaping. 1285 The need for a lower-level DetNet node to be aware of individual 1286 higher-layer flows is not unique to DetNet. But, given the endless 1287 complexity of layering and relayering over tunnels that is available 1288 to network designers, DetNet needs to provide a model for flow 1289 identification that is at least somewhat better than packet 1290 inspection. That is not to say that packet inspection to layer 4 or 1291 5 addresses will not be used, or the capability standardized; but, 1292 there are alternatives. 1294 A DetNet relay node can connect DetNet flows on different paths using 1295 different flow identification methods. For example: 1297 o A single unicast DetNet flow passing from router A through a 1298 bridged network to router B may be assigned a {VLAN, multicast 1299 destination MAC address} pair that is unique within that bridged 1300 network. The bridges can then identify the flow without accessing 1301 higher-layer headers. Of course, the receiving router must 1302 recognize and accept that multicast MAC address. 1304 o A DetNet flow passing from LSR A to LSR B may be assigned a 1305 different label than that used for other flows to the same IP 1306 destination. 1308 In any of the above cases, it is possible that an existing DetNet 1309 flow can be used as a carrier for multiple DetNet sub-flows. (Not to 1310 be confused with DetNet compound vs. member flows.) Of course, this 1311 requires that the aggregate DetNet flow be provisioned properly to 1312 carry the sub-flows. 1314 Thus, rather than packet inspection, there is the option to export 1315 higher-layer information to the lower layer. The requirement to 1316 support one or the other method for flow identification (or both) is 1317 the essential complexity that DetNet brings to existing control plane 1318 models. 1320 4.7.2. Flow attribute mapping between layers 1322 Transport of DetNet flows over multiple technology domains may 1323 require that lower layers are aware of specific flows of higher 1324 layers. Such an "exporting of flow identification" is needed each 1325 time when the forwarding paradigm is changed on the transport path 1326 (e.g., two LSRs are interconnected by a L2 bridged domain, etc.). 1327 The three main forwarding methods considered for deterministic 1328 networking are: 1330 o IP routing 1332 o MPLS label switching 1334 o Ethernet bridging 1336 Note: at the time of this publication, the exact format of flow 1337 identification is still WIP. 1339 [Note: Seq-num attribute may require a similar functionality at 1340 technology border nodes.] 1342 add/remove add/remove 1343 Eth Flow-ID IP Flow-ID 1344 | | 1345 v v 1346 +-----------------------------------------------------------+ 1347 | | | | | 1348 | Eth | MPLS | IP | Application data | 1349 | | | | | 1350 +-----------------------------------------------------------+ 1351 ^ 1352 | 1353 add/remove 1354 MPLS Flow-ID 1356 Figure 9: Packet with multiple Flow-IDs 1358 The additional (domain specific) Flow-ID can be 1360 o created by a domain specific function or 1362 o derived from the Flow-ID added to the App-flow, 1364 so that it must be unique inside the given domain. Note, that the 1365 Flow-ID added to the App-flow is still present in the packet, but 1366 transport nodes may lack the function to recognize it; that's why the 1367 additional Flow-ID is added (pushed). 1369 4.7.3. Flow-ID mapping examples 1371 IP nodes and MPLS nodes are assumed to be configured to push such an 1372 additional (domain specific) Flow-ID when sending traffic to an 1373 Ethernet switch (as shown in the examples below). 1375 Figure 10 shows a scenario where an IP end system ("IP-A") is 1376 connected via two Ethernet switches ("ETH-n") to an IP router ("IP- 1377 1"). 1379 IP domain 1380 <----------------------------------------------- 1382 +======+ +======+ 1383 |L3-ID | |L3-ID | 1384 +======+ /\ +-----+ +======+ 1385 / \ Forward as | | 1386 /IP-A\ per ETH-ID |IP-1 | Recognize 1387 Push ------> +-+----+ | +---+-+ <----- ETH-ID 1388 ETH-ID | +----+-----+ | 1389 | v v | 1390 | +-----+ +-----+ | 1391 +------+ | | +---------+ 1392 +......+ |ETH-1+----+ETH-2| +======+ 1393 .L3-ID . +-----+ +-----+ |L3-ID | 1394 +======+ +......+ +======+ 1395 |ETH-ID| .L3-ID . |ETH-ID| 1396 +======+ +======+ +------+ 1397 |ETH-ID| 1398 +======+ 1400 Ethernet domain 1401 <----------------> 1403 Figure 10: IP nodes interconnected by an Ethernet domain 1405 End system "IP-A" uses the original App-flow specific ID ("L3-ID"), 1406 but as it is connected to an Ethernet domain it has to push an 1407 Ethernet-domain specific flow-ID ("VID + multicast MAC address", 1408 referred as "ETH-ID") before sending the packet to "ETH-1" node. 1409 Ethernet switch "ETH-1" can recognize the data flow based on the 1410 "ETH-ID" and it does forwarding toward "ETH-2". "ETH-2" switches the 1411 packet toward the IP router. "IP-1" must be configured to receive 1412 the Ethernet Flow-ID specific multicast stream, but (as it is an L3 1413 node) it decodes the data flow ID based on the "L3-ID" fields of the 1414 received packet. 1416 Figure 11 shows a scenario where MPLS domain nodes ("PE-n" and "P-m") 1417 are connected via two Ethernet switches ("ETH-n"). 1419 MPLS domain 1420 <-----------------------------------------------> 1422 +=======+ +=======+ 1423 |MPLS-ID| |MPLS-ID| 1424 +=======+ +-----+ +-----+ +=======+ +-----+ 1425 | | Forward as | | | | 1426 |PE-1 | per ETH-ID | P-2 +-----------+ PE-2| 1427 Push -----> +-+---+ | +---+-+ +-----+ 1428 ETH-ID | +-----+----+ | \ Recognize 1429 | v v | +-- ETH-ID 1430 | +-----+ +-----+ | 1431 +---+ | | +----+ 1432 +.......+ |ETH-1+----+ETH-2| +=======+ 1433 .MPLS-ID. +-----+ +-----+ |MPLS-ID| 1434 +=======+ +=======+ 1435 |ETH-ID | +.......+ |ETH-ID | 1436 +=======+ .MPLS-ID. +-------+ 1437 +=======+ 1438 |ETH-ID | 1439 +=======+ 1440 Ethernet domain 1441 <----------------> 1443 Figure 11: MPLS nodes interconnected by an Ethernet domain 1445 "PE-1" uses the MPLS specific ID ("MPLS-ID"), but as it is connected 1446 to an Ethernet domain it has to push an Ethernet-domain specific 1447 flow-ID ("VID + multicast MAC address", referred as "ETH-ID") before 1448 sending the packet to "ETH-1". Ethernet switch "ETH-1" can recognize 1449 the data flow based on the "ETH-ID" and it does forwarding toward 1450 "ETH-2". "ETH-2" switches the packet toward the MPLS node ("P-2"). 1451 "P-2" must be configured to receive the Ethernet Flow-ID specific 1452 multicast stream, but (as it is an MPLS node) it decodes the data 1453 flow ID based on the "MPLS-ID" fields of the received packet. 1455 4.8. Advertising resources, capabilities and adjacencies 1457 There are three classes of information that a central controller or 1458 decentralized control plane needs to know that can only be obtained 1459 from the end systems and/or transit nodes in the network. When using 1460 a peer-to-peer control plane, some of this information may be 1461 required by a system's neighbors in the network. 1463 o Details of the system's capabilities that are required in order to 1464 accurately allocate that system's resources, as well as other 1465 systems' resources. This includes, for example, which specific 1466 queuing and shaping algorithms are implemented (Section 4.5), the 1467 number of buffers dedicated for DetNet allocation, and the worst- 1468 case forwarding delay. 1470 o The dynamic state of an end or transit node's DetNet resources. 1472 o The identity of the system's neighbors, and the characteristics of 1473 the link(s) between the systems, including the length (in 1474 nanoseconds) of the link(s). 1476 4.9. Provisioning model 1478 4.9.1. Centralized Path Computation and Installation 1480 A centralized routing model, such as provided with a PCE (RFC 4655 1481 [RFC4655]), enables global and per-flow optimizations. (See 1482 Section 4.4.) The model is attractive but a number of issues are 1483 left to be solved. In particular: 1485 o Whether and how the path computation can be installed by 1) an end 1486 device or 2) a Network Management entity, 1488 o And how the path is set up, either by installing state at each hop 1489 with a direct interaction between the forwarding device and the 1490 PCE, or along a path by injecting a source-routed request at one 1491 end of the path. 1493 4.9.2. Distributed Path Setup 1495 Significant work on distributed path setup can be leveraged from MPLS 1496 Traffic Engineering, in both its GMPLS and non-GMPLS forms. The 1497 protocols within scope are Resource ReSerVation Protocol [RFC3209] 1498 [RFC3473](RSVP-TE), OSPF-TE [RFC4203] [RFC5392] and ISIS-TE [RFC5307] 1500 [RFC5316]. These should be viewed as starting points as there are 1501 feature specific extensions defined that may be applicable to DetNet. 1503 In a Layer-2 only environment, or as part of a layered approach to a 1504 mixed environment, IEEE 802.1 also has work, either completed or in 1505 progress. [IEEE802.1Q-2014] Clause 35 describes SRP, a peer-to-peer 1506 protocol for Layer-2 roughly analogous to RSVP [RFC2205]. 1507 [IEEE802.1Qca] defines how ISIS can provide multiple disjoint paths 1508 or distribution trees. Also in progress is [IEEE802.1Qcc], which 1509 expands the capabilities of SRP. 1511 The integration/interaction of the DetNet control layer with an 1512 underlying IEEE 802.1 sub-network control layer will need to be 1513 defined. 1515 4.10. Scaling to larger networks 1517 Reservations for individual DetNet flows require considerable state 1518 information in each transit node, especially when adequate fault 1519 mitigation (Section 3.3.2) is required. The DetNet data plane, in 1520 order to support larger numbers of DetNet flows, must support the 1521 aggregation of DetNet flows into tunnels, which themselves can be 1522 viewed by the transit nodes' data planes largely as individual DetNet 1523 flows. Without such aggregation, the per-relay system may limit the 1524 scale of DetNet networks. 1526 4.11. Connected islands vs. networks 1528 Given that users have deployed examples of the IEEE 802.1 TSN TG 1529 standards, which provide capabilities similar to DetNet, it is 1530 obvious to ask whether the IETF DetNet effort can be limited to 1531 providing Layer-2 connections (VPNs) between islands of bridged TSN 1532 networks. While this capability is certainly useful to some 1533 applications, and must not be precluded by DetNet, tunneling alone is 1534 not a sufficient goal for the DetNet WG. As shown in the 1535 Deterministic Networking Use Cases draft [I-D.ietf-detnet-use-cases], 1536 there are already deployments of Layer-2 TSN networks that are 1537 encountering the well-known problems of over-large broadcast domains. 1538 Routed solutions, and combinations routed/bridged solutions, are both 1539 required. 1541 4.12. Compatibility with Layer-2 1543 Standards providing similar capabilities for bridged networks (only) 1544 have been and are being generated in the IEEE 802 LAN/MAN Standards 1545 Committee. The present architecture describes an abstract model that 1546 can be applicable both at Layer-2 and Layer-3, and over links not 1547 defined by IEEE 802. It is the intention of the authors (and 1548 hopefully, as this draft progresses, of the DetNet Working Group) 1549 that IETF and IEEE 802 will coordinate their work, via the 1550 participation of common individuals, liaisons, and other means, to 1551 maximize the compatibility of their outputs. 1553 DetNet enabled end systems and intermediate nodes can be 1554 interconnected by sub-networks, i.e., Layer-2 technologies. These 1555 sub-networks will provide DetNet compatible service for support of 1556 DetNet traffic. Examples of sub-networks include 802.1TSN and a 1557 point-to-point OTN link. Of course, multi-layer DetNet systems may 1558 be possible too, where one DetNet appears as a sub-network, and 1559 provides service to, a higher layer DetNet system. 1561 5. Open Questions 1563 There are a number of architectural questions that will have to be 1564 resolved before this document can be submitted for publication. 1565 Aside from the obvious fact that this present draft is subject to 1566 change, there are specific questions to which the authors wish to 1567 direct the readers' attention. 1569 5.1. Flat vs. hierarchical control 1571 Boxes that are solely routers or solely bridges are rare in today's 1572 market. In a multi-tenant data center, multiple users' virtual 1573 Layer-2/Layer-3 topologies exist simultaneously, implemented on a 1574 network whose physical topology bears only accidental resemblance to 1575 the virtual topologies. 1577 While the forwarding topology (the bridges and routers) are an 1578 important consideration for a DetNet Flow Management Entity 1579 (Section 4.4.1), so is the purely physical topology. Ultimately, the 1580 model used by the management entities is based on boxes, queues, and 1581 links. The authors hope that the work of the TEAS WG will help to 1582 clarify exactly what model parameters need to be traded between the 1583 intermediate nodes and the controller(s). 1585 5.2. Peer-to-peer reservation protocol 1587 As described in Section 4.9.2, the DetNet WG needs to decide whether 1588 to support a peer-to-peer protocol for a source and a destination to 1589 reserve resources for a DetNet stream. Assuming that enabling the 1590 involvement of the source and/or destination is desirable (see 1591 Deterministic Networking Use Cases [I-D.ietf-detnet-use-cases]), it 1592 remains to decide whether the DetNet WG will make it possible to 1593 deploy at least some DetNet capabilities in a network using only a 1594 peer-to-peer protocol, without a central controller. 1596 (Note that a UNI (see Section 4.4.3) between an end system and a 1597 DetNet edge node, for sources and/or listeners to request DetNet 1598 services, can be either the first hop of a per-to-peer reservation 1599 protocol, or can be deflected by the DetNet edge node to a central 1600 controller for resolution. Similarly, a decision by a central 1601 controller can be effected by the controller instructing the end 1602 system or DetNet edge node to initiate a per-to-peer protocol 1603 activity.) 1605 5.3. Wireless media interactions 1607 Deterministic Networking Use Cases [I-D.ietf-detnet-use-cases] 1608 illustrates cases where wireless media are needed in a DetNet 1609 network. Some wireless media in general use, such as IEEE 802.11 1610 [IEEE802.1Q-2014], have significantly higher packet loss rates than 1611 typical wired media, such as Ethernet [IEEE802.3-2012]. IEEE 802.11 1612 includes support for such features as MAC-layer acknowledgements and 1613 retransmissions. 1615 The techniques described in Section 3 are likely to improve the 1616 ability of a mixed wired/wireless network to offer the DetNet QoS 1617 features. The interaction of these techniques with the features of 1618 specific wireless media, although they may be significant, cannot be 1619 addressed in this document. It remains to be decided to what extent 1620 the DetNet WG will address them, and to what extent other WGs, e.g. 1621 6TiSCH, will do so. 1623 5.4. Packet encoding for service protection 1625 There are methods for using multiple paths to provide service 1626 protection that involve encoding the information in a packet 1627 belonging to a DetNet flow into multiple transmission units, 1628 typically combining information from multiple packets into any given 1629 transmission unit. Such techniques may be applicable for use as a 1630 DetNet service protection technique, assuming that the DetNet users' 1631 needs for timeliness of delivery and freedom from interference with 1632 misbehaving DetNet flows can be met. 1634 No specific mechanisms are defined here, at this time. This section 1635 will either be enhanced or removed. Contributions are invited. 1637 6. Security Considerations 1639 Security in the context of Deterministic Networking has an added 1640 dimension; the time of delivery of a packet can be just as important 1641 as the contents of the packet, itself. A man-in-the-middle attack, 1642 for example, can impose, and then systematically adjust, additional 1643 delays into a link, and thus disrupt or subvert a real-time 1644 application without having to crack any encryption methods employed. 1645 See [RFC7384] for an exploration of this issue in a related context. 1647 Furthermore, in a control system where millions of dollars of 1648 equipment, or even human lives, can be lost if the DetNet QoS is not 1649 delivered, one must consider not only simple equipment failures, 1650 where the box or wire instantly becomes perfectly silent, but bizarre 1651 errors such as can be caused by software failures. Because there is 1652 essential no limit to the kinds of failures that can occur, 1653 protecting against realistic equipment failures is indistinguishable, 1654 in most cases, from protecting against malicious behavior, whether 1655 accidental or intentional. See also Section 3.3.2. 1657 Security must cover: 1659 o the protection of the signaling protocol 1661 o the authentication and authorization of the controlling systems 1663 o the identification and shaping of the DetNet flows 1665 7. Privacy Considerations 1667 DetNet is provides a Quality of Service (QoS), and as such, does not 1668 directly raise any new privacy considerations. 1670 However, the requirement for every (or almost every) node along the 1671 path of a DetNet flow to identify DetNet flows may present an 1672 additional attack surface for privacy, should the DetNet paradigm be 1673 found useful in broader environments. 1675 8. IANA Considerations 1677 This document does not require an action from IANA. 1679 9. Acknowledgements 1681 The authors wish to thank Jouni Korhonen, Erik Nordmark, George 1682 Swallow, Rudy Klecka, Anca Zamfir, David Black, Thomas Watteyne, 1683 Shitanshu Shah, Craig Gunther, Rodney Cummings, Balazs Varga, 1684 Wilfried Steiner, Marcel Kiessling, Karl Weber, Janos Farkas, Ethan 1685 Grossman, Pat Thaler, Lou Berger, and especially Michael Johas 1686 Teener, for their various contribution with this work. 1688 10. Access to IEEE 802.1 documents 1690 To access password protected IEEE 802.1 drafts, see the IETF IEEE 1691 802.1 information page at https://www.ietf.org/proceedings/52/slides/ 1692 bridge-0/tsld003.htm. 1694 11. Informative References 1696 [AVnu] http://www.avnu.org/, "The AVnu Alliance tests and 1697 certifies devices for interoperability, providing a simple 1698 and reliable networking solution for AV network 1699 implementation based on the Audio Video Bridging (AVB) 1700 standards.". 1702 [CCAMP] IETF, "Common Control and Measurement Plane", 1703 . 1705 [HART] www.hartcomm.org, "Highway Addressable Remote Transducer, 1706 a group of specifications for industrial process and 1707 control devices administered by the HART Foundation". 1709 [HSR-PRP] IEC, "High availability seamless redundancy (HSR) is a 1710 further development of the PRP approach, although HSR 1711 functions primarily as a protocol for creating media 1712 redundancy while PRP, as described in the previous 1713 section, creates network redundancy. PRP and HSR are both 1714 described in the IEC 62439 3 standard.", 1715 . 1718 [I-D.dt-detnet-dp-alt] 1719 Korhonen, J., Farkas, J., Mirsky, G., Thubert, P., 1720 Zhuangyan, Z., and L. Berger, "DetNet Data Plane Protocol 1721 and Solution Alternatives", draft-dt-detnet-dp-alt-04 1722 (work in progress), September 2016. 1724 [I-D.ietf-6tisch-architecture] 1725 Thubert, P., "An Architecture for IPv6 over the TSCH mode 1726 of IEEE 802.15.4", draft-ietf-6tisch-architecture-11 (work 1727 in progress), January 2017. 1729 [I-D.ietf-6tisch-tsch] 1730 Watteyne, T., Palattella, M., and L. Grieco, "Using 1731 IEEE802.15.4e TSCH in an IoT context: Overview, Problem 1732 Statement and Goals", draft-ietf-6tisch-tsch-06 (work in 1733 progress), March 2015. 1735 [I-D.ietf-detnet-problem-statement] 1736 Finn, N. and P. Thubert, "Deterministic Networking Problem 1737 Statement", draft-ietf-detnet-problem-statement-01 (work 1738 in progress), September 2016. 1740 [I-D.ietf-detnet-use-cases] 1741 Grossman, E., Gunther, C., Thubert, P., Wetterwald, P., 1742 Raymond, J., Korhonen, J., Kaneko, Y., Das, S., Zha, Y., 1743 Varga, B., Farkas, J., Goetz, F., Schmitt, J., Vilajosana, 1744 X., Mahmoodi, T., Spirou, S., and P. Vizarreta, 1745 "Deterministic Networking Use Cases", draft-ietf-detnet- 1746 use-cases-12 (work in progress), April 2017. 1748 [I-D.ietf-roll-rpl-industrial-applicability] 1749 Phinney, T., Thubert, P., and R. Assimiti, "RPL 1750 applicability in industrial networks", draft-ietf-roll- 1751 rpl-industrial-applicability-02 (work in progress), 1752 October 2013. 1754 [I-D.svshah-tsvwg-deterministic-forwarding] 1755 Shah, S. and P. Thubert, "Deterministic Forwarding PHB", 1756 draft-svshah-tsvwg-deterministic-forwarding-04 (work in 1757 progress), August 2015. 1759 [I-D.varga-detnet-service-model] 1760 Varga, B. and J. Farkas, "DetNet Service Model", draft- 1761 varga-detnet-service-model-02 (work in progress), May 1762 2017. 1764 [IEEE802.11-2012] 1765 IEEE, "Wireless LAN Medium Access Control (MAC) and 1766 Physical Layer (PHY) Specifications", 2012, 1767 . 1770 [IEEE802.1AS-2011] 1771 IEEE, "Timing and Synchronizations (IEEE 802.1AS-2011)", 1772 2011, . 1775 [IEEE802.1BA-2011] 1776 IEEE, "AVB (Audio Video Bridging) Systems (IEEE 802.1BA- 1777 2011)", 2011, . 1780 [IEEE802.1CB] 1781 IEEE, "Frame Replication and Elimination for Reliability 1782 (IEEE Draft P802.1CB)", 2016, 1783 . 1785 [IEEE802.1Q-2014] 1786 IEEE, "MAC Bridges and VLANs (IEEE 802.1Q-2014", 2014, 1787 . 1790 [IEEE802.1Qbu] 1791 IEEE, "Frame Preemption", 2016, 1792 . 1794 [IEEE802.1Qbv] 1795 IEEE, "Enhancements for Scheduled Traffic", 2016, 1796 . 1798 [IEEE802.1Qca] 1799 IEEE 802.1, "IEEE 802.1Qca Bridges and Bridged Networks - 1800 Amendment 24: Path Control and Reservation", IEEE 1801 P802.1Qca/D2.1 P802.1Qca, June 2015, 1802 . 1805 [IEEE802.1Qcc] 1806 IEEE, "Stream Reservation Protocol (SRP) Enhancements and 1807 Performance Improvements", 2016, 1808 . 1810 [IEEE802.1Qch] 1811 IEEE, "Cyclic Queuing and Forwarding", 2016, 1812 . 1814 [IEEE802.1TSNTG] 1815 IEEE Standards Association, "IEEE 802.1 Time-Sensitive 1816 Networks Task Group", 2013, 1817 . 1819 [IEEE802.3-2012] 1820 IEEE, "IEEE Standard for Ethernet", 2012, 1821 . 1824 [IEEE802.3br] 1825 IEEE, "Interspersed Express Traffic", 2016, 1826 . 1828 [IEEE802154] 1829 IEEE Standard for Information Technology, "IEEE 802.15.4, 1830 Part. 15.4: Wireless Medium Access Control (MAC) and 1831 Physical Layer (PHY) Specifications for Low-Rate Wireless 1832 Personal Area Networks", June 2011. 1834 [IEEE802154e] 1835 IEEE Standard for Information Technology, "IEEE 802.15.4e, 1836 Part. 15.4: Low-Rate Wireless Personal Area Networks (LR- 1837 WPANs) Amendment 1: MAC sublayer", April 2012. 1839 [ISA100.11a] 1840 ISA/IEC, "ISA100.11a, Wireless Systems for Automation, 1841 also IEC 62734", 2011, < http://www.isa100wci.org/en- 1842 US/Documents/PDF/3405-ISA100-WirelessSystems-Future-broch- 1843 WEB-ETSI.aspx>. 1845 [ISA95] ANSI/ISA, "Enterprise-Control System Integration Part 1: 1846 Models and Terminology", 2000, . 1849 [ODVA] http://www.odva.org/, "The organization that supports 1850 network technologies built on the Common Industrial 1851 Protocol (CIP) including EtherNet/IP.". 1853 [PCE] IETF, "Path Computation Element", 1854 . 1856 [Profinet] 1857 http://us.profinet.com/technology/profinet/, "PROFINET is 1858 a standard for industrial networking in automation.", 1859 . 1861 [RFC2205] Braden, R., Ed., Zhang, L., Berson, S., Herzog, S., and S. 1862 Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1 1863 Functional Specification", RFC 2205, DOI 10.17487/RFC2205, 1864 September 1997, . 1866 [RFC2475] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z., 1867 and W. Weiss, "An Architecture for Differentiated 1868 Services", RFC 2475, DOI 10.17487/RFC2475, December 1998, 1869 . 1871 [RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V., 1872 and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP 1873 Tunnels", RFC 3209, DOI 10.17487/RFC3209, December 2001, 1874 . 1876 [RFC3473] Berger, L., Ed., "Generalized Multi-Protocol Label 1877 Switching (GMPLS) Signaling Resource ReserVation Protocol- 1878 Traffic Engineering (RSVP-TE) Extensions", RFC 3473, 1879 DOI 10.17487/RFC3473, January 2003, 1880 . 1882 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 1883 Jacobson, "RTP: A Transport Protocol for Real-Time 1884 Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550, 1885 July 2003, . 1887 [RFC4203] Kompella, K., Ed. and Y. Rekhter, Ed., "OSPF Extensions in 1888 Support of Generalized Multi-Protocol Label Switching 1889 (GMPLS)", RFC 4203, DOI 10.17487/RFC4203, October 2005, 1890 . 1892 [RFC4655] Farrel, A., Vasseur, J., and J. Ash, "A Path Computation 1893 Element (PCE)-Based Architecture", RFC 4655, 1894 DOI 10.17487/RFC4655, August 2006, 1895 . 1897 [RFC5307] Kompella, K., Ed. and Y. Rekhter, Ed., "IS-IS Extensions 1898 in Support of Generalized Multi-Protocol Label Switching 1899 (GMPLS)", RFC 5307, DOI 10.17487/RFC5307, October 2008, 1900 . 1902 [RFC5316] Chen, M., Zhang, R., and X. Duan, "ISIS Extensions in 1903 Support of Inter-Autonomous System (AS) MPLS and GMPLS 1904 Traffic Engineering", RFC 5316, DOI 10.17487/RFC5316, 1905 December 2008, . 1907 [RFC5392] Chen, M., Zhang, R., and X. Duan, "OSPF Extensions in 1908 Support of Inter-Autonomous System (AS) MPLS and GMPLS 1909 Traffic Engineering", RFC 5392, DOI 10.17487/RFC5392, 1910 January 2009, . 1912 [RFC5673] Pister, K., Ed., Thubert, P., Ed., Dwars, S., and T. 1913 Phinney, "Industrial Routing Requirements in Low-Power and 1914 Lossy Networks", RFC 5673, DOI 10.17487/RFC5673, October 1915 2009, . 1917 [RFC5921] Bocci, M., Ed., Bryant, S., Ed., Frost, D., Ed., Levrau, 1918 L., and L. Berger, "A Framework for MPLS in Transport 1919 Networks", RFC 5921, DOI 10.17487/RFC5921, July 2010, 1920 . 1922 [RFC6372] Sprecher, N., Ed. and A. Farrel, Ed., "MPLS Transport 1923 Profile (MPLS-TP) Survivability Framework", RFC 6372, 1924 DOI 10.17487/RFC6372, September 2011, 1925 . 1927 [RFC6658] Bryant, S., Ed., Martini, L., Swallow, G., and A. Malis, 1928 "Packet Pseudowire Encapsulation over an MPLS PSN", 1929 RFC 6658, DOI 10.17487/RFC6658, July 2012, 1930 . 1932 [RFC7384] Mizrahi, T., "Security Requirements of Time Protocols in 1933 Packet Switched Networks", RFC 7384, DOI 10.17487/RFC7384, 1934 October 2014, . 1936 [RFC7426] Haleplidis, E., Ed., Pentikousis, K., Ed., Denazis, S., 1937 Hadi Salim, J., Meyer, D., and O. Koufopavlou, "Software- 1938 Defined Networking (SDN): Layers and Architecture 1939 Terminology", RFC 7426, DOI 10.17487/RFC7426, January 1940 2015, . 1942 [TEAS] IETF, "Traffic Engineering Architecture and Signaling", 1943 . 1945 [WirelessHART] 1946 www.hartcomm.org, "Industrial Communication Networks - 1947 Wireless Communication Network and Communication Profiles 1948 - WirelessHART - IEC 62591", 2010. 1950 Authors' Addresses 1952 Norman Finn 1953 Huawei Technologies Co. Ltd 1954 3755 Avocado Blvd. 1955 PMB 436 1956 La Mesa, California 91941 1957 US 1959 Phone: +1 925 980 6430 1960 Email: norman.finn@mail01.huawei.com 1961 Pascal Thubert 1962 Cisco Systems 1963 Village d'Entreprises Green Side 1964 400, Avenue de Roumanille 1965 Batiment T3 1966 Biot - Sophia Antipolis 06410 1967 FRANCE 1969 Phone: +33 4 97 23 26 34 1970 Email: pthubert@cisco.com 1972 Balazs Varga 1973 Ericsson 1974 Konyves Kalman krt. 11/B 1975 Budapest 1097 1976 Hungary 1978 Email: balazs.a.varga@ericsson.com 1980 Janos Farkas 1981 Ericsson 1982 Konyves Kalman krt. 11/B 1983 Budapest 1097 1984 Hungary 1986 Email: janos.farkas@ericsson.com