idnits 2.17.1 draft-ietf-detnet-use-cases-13.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- Couldn't find a document date in the document -- date freshness check skipped. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'ACE' is defined on line 3514, but no explicit reference was found in the text == Unused Reference: 'CCAMP' is defined on line 3526, but no explicit reference was found in the text == Unused Reference: 'CPRI-transp' is defined on line 3545, but no explicit reference was found in the text == Unused Reference: 'DICE' is defined on line 3554, but no explicit reference was found in the text == Unused Reference: 'EA12' is defined on line 3557, but no explicit reference was found in the text == Unused Reference: 'HART' is defined on line 3574, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-6tisch-terminology' is defined on line 3603, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-ipv6-multilink-subnets' is defined on line 3609, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-roll-rpl-industrial-applicability' is defined on line 3620, but no explicit reference was found in the text == Unused Reference: 'I-D.thubert-6lowpan-backbone-router' is defined on line 3642, but no explicit reference was found in the text == Unused Reference: 'IEC61850-90-12' is defined on line 3661, but no explicit reference was found in the text == Unused Reference: 'IEEE8021TSN' is defined on line 3729, but no explicit reference was found in the text == Unused Reference: 'IETFDetNet' is defined on line 3735, but no explicit reference was found in the text == Unused Reference: 'ISO7240-16' is defined on line 3748, but no explicit reference was found in the text == Unused Reference: 'LTE-Latency' is defined on line 3759, but no explicit reference was found in the text == Unused Reference: 'RFC2119' is defined on line 3803, but no explicit reference was found in the text == Unused Reference: 'RFC2460' is defined on line 3808, but no explicit reference was found in the text == Unused Reference: 'RFC2474' is defined on line 3812, but no explicit reference was found in the text == Unused Reference: 'RFC3209' is defined on line 3823, but no explicit reference was found in the text == Unused Reference: 'RFC3393' is defined on line 3828, but no explicit reference was found in the text == Unused Reference: 'RFC3444' is defined on line 3839, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3844, but no explicit reference was found in the text == Unused Reference: 'RFC4291' is defined on line 3853, but no explicit reference was found in the text == Unused Reference: 'RFC4903' is defined on line 3862, but no explicit reference was found in the text == Unused Reference: 'RFC4919' is defined on line 3866, but no explicit reference was found in the text == Unused Reference: 'RFC6282' is defined on line 3883, but no explicit reference was found in the text == Unused Reference: 'RFC6775' is defined on line 3901, but no explicit reference was found in the text == Unused Reference: 'TEAS' is defined on line 3932, but no explicit reference was found in the text == Unused Reference: 'UHD-video' is defined on line 3963, but no explicit reference was found in the text == Outdated reference: A later version (-30) exists of draft-ietf-6tisch-architecture-12 == Outdated reference: A later version (-10) exists of draft-ietf-6tisch-terminology-09 -- Obsolete informational reference (is this intentional?): RFC 2460 (Obsoleted by RFC 8200) Summary: 1 error (**), 0 flaws (~~), 32 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Engineering Task Force E. Grossman, Ed. 3 Internet-Draft DOLBY 4 Intended status: Informational C. Gunther 5 Expires: March 22, 2018 HARMAN 6 P. Thubert 7 P. Wetterwald 8 CISCO 9 J. Raymond 10 HYDRO-QUEBEC 11 J. Korhonen 12 BROADCOM 13 Y. Kaneko 14 Toshiba 15 S. Das 16 Applied Communication Sciences 17 Y. Zha 18 HUAWEI 19 B. Varga 20 J. Farkas 21 Ericsson 22 F. Goetz 23 J. Schmitt 24 Siemens 25 X. Vilajosana 26 Worldsensing 27 T. Mahmoodi 28 King's College London 29 S. Spirou 30 Intracom Telecom 31 P. Vizarreta 32 Technical University of Munich, TUM 33 D. Huang 34 ZTE Corporation, Inc. 35 X. Geng 36 HUAWEI 37 D. Dujovne 38 UDP 39 M. Seewald 40 CISCO 41 September 18, 2017 43 Deterministic Networking Use Cases 44 draft-ietf-detnet-use-cases-13 46 Abstract 48 This draft documents requirements in several diverse industries to 49 establish multi-hop paths for characterized flows with deterministic 50 properties. In this context deterministic implies that streams can 51 be established which provide guaranteed bandwidth and latency which 52 can be established from either a Layer 2 or Layer 3 (IP) interface, 53 and which can co-exist on an IP network with best-effort traffic. 55 Additional requirements include optional redundant paths, very high 56 reliability paths, time synchronization, and clock distribution. 57 Industries considered include professional audio, electrical 58 utilities, building automation systems, wireless for industrial, 59 cellular radio, industrial machine-to-machine, mining, private 60 blockchain, and network slicing. 62 For each case, this document will identify the application, identify 63 representative solutions used today, and the improvements IETF DetNet 64 solutions may enable. 66 Status of This Memo 68 This Internet-Draft is submitted in full conformance with the 69 provisions of BCP 78 and BCP 79. 71 Internet-Drafts are working documents of the Internet Engineering 72 Task Force (IETF). Note that other groups may also distribute 73 working documents as Internet-Drafts. The list of current Internet- 74 Drafts is at https://datatracker.ietf.org/drafts/current/. 76 Internet-Drafts are draft documents valid for a maximum of six months 77 and may be updated, replaced, or obsoleted by other documents at any 78 time. It is inappropriate to use Internet-Drafts as reference 79 material or to cite them other than as "work in progress." 81 This Internet-Draft will expire on March 22, 2018. 83 Copyright Notice 85 Copyright (c) 2017 IETF Trust and the persons identified as the 86 document authors. All rights reserved. 88 This document is subject to BCP 78 and the IETF Trust's Legal 89 Provisions Relating to IETF Documents 90 (https://trustee.ietf.org/license-info) in effect on the date of 91 publication of this document. Please review these documents 92 carefully, as they describe your rights and restrictions with respect 93 to this document. Code Components extracted from this document must 94 include Simplified BSD License text as described in Section 4.e of 95 the Trust Legal Provisions and are provided without warranty as 96 described in the Simplified BSD License. 98 Table of Contents 100 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 6 101 2. Pro Audio and Video . . . . . . . . . . . . . . . . . . . . . 7 102 2.1. Use Case Description . . . . . . . . . . . . . . . . . . 7 103 2.1.1. Uninterrupted Stream Playback . . . . . . . . . . . . 8 104 2.1.2. Synchronized Stream Playback . . . . . . . . . . . . 8 105 2.1.3. Sound Reinforcement . . . . . . . . . . . . . . . . . 9 106 2.1.4. Deterministic Time to Establish Streaming . . . . . . 9 107 2.1.5. Secure Transmission . . . . . . . . . . . . . . . . . 9 108 2.1.5.1. Safety . . . . . . . . . . . . . . . . . . . . . 9 109 2.2. Pro Audio Today . . . . . . . . . . . . . . . . . . . . . 9 110 2.3. Pro Audio Future . . . . . . . . . . . . . . . . . . . . 10 111 2.3.1. Layer 3 Interconnecting Layer 2 Islands . . . . . . . 10 112 2.3.2. High Reliability Stream Paths . . . . . . . . . . . . 10 113 2.3.3. Integration of Reserved Streams into IT Networks . . 10 114 2.3.4. Use of Unused Reservations by Best-Effort Traffic . . 11 115 2.3.5. Traffic Segregation . . . . . . . . . . . . . . . . . 11 116 2.3.5.1. Packet Forwarding Rules, VLANs and Subnets . . . 11 117 2.3.5.2. Multicast Addressing (IPv4 and IPv6) . . . . . . 12 118 2.3.6. Latency Optimization by a Central Controller . . . . 12 119 2.3.7. Reduced Device Cost Due To Reduced Buffer Memory . . 12 120 2.4. Pro Audio Asks . . . . . . . . . . . . . . . . . . . . . 13 121 3. Electrical Utilities . . . . . . . . . . . . . . . . . . . . 13 122 3.1. Use Case Description . . . . . . . . . . . . . . . . . . 13 123 3.1.1. Transmission Use Cases . . . . . . . . . . . . . . . 13 124 3.1.1.1. Protection . . . . . . . . . . . . . . . . . . . 13 125 3.1.1.2. Intra-Substation Process Bus Communications . . . 19 126 3.1.1.3. Wide Area Monitoring and Control Systems . . . . 20 127 3.1.1.4. IEC 61850 WAN engineering guidelines requirement 128 classification . . . . . . . . . . . . . . . . . 21 129 3.1.2. Generation Use Case . . . . . . . . . . . . . . . . . 22 130 3.1.2.1. Control of the Generated Power . . . . . . . . . 22 131 3.1.2.2. Control of the Generation Infrastructure . . . . 23 132 3.1.3. Distribution use case . . . . . . . . . . . . . . . . 28 133 3.1.3.1. Fault Location Isolation and Service Restoration 134 (FLISR) . . . . . . . . . . . . . . . . . . . . . 28 135 3.2. Electrical Utilities Today . . . . . . . . . . . . . . . 29 136 3.2.1. Security Current Practices and Limitations . . . . . 29 137 3.3. Electrical Utilities Future . . . . . . . . . . . . . . . 31 138 3.3.1. Migration to Packet-Switched Network . . . . . . . . 32 139 3.3.2. Telecommunications Trends . . . . . . . . . . . . . . 32 140 3.3.2.1. General Telecommunications Requirements . . . . . 32 141 3.3.2.2. Specific Network topologies of Smart Grid 142 Applications . . . . . . . . . . . . . . . . . . 33 143 3.3.2.3. Precision Time Protocol . . . . . . . . . . . . . 34 144 3.3.3. Security Trends in Utility Networks . . . . . . . . . 35 145 3.4. Electrical Utilities Asks . . . . . . . . . . . . . . . . 37 146 4. Building Automation Systems . . . . . . . . . . . . . . . . . 37 147 4.1. Use Case Description . . . . . . . . . . . . . . . . . . 37 148 4.2. Building Automation Systems Today . . . . . . . . . . . . 38 149 4.2.1. BAS Architecture . . . . . . . . . . . . . . . . . . 38 150 4.2.2. BAS Deployment Model . . . . . . . . . . . . . . . . 39 151 4.2.3. Use Cases for Field Networks . . . . . . . . . . . . 41 152 4.2.3.1. Environmental Monitoring . . . . . . . . . . . . 41 153 4.2.3.2. Fire Detection . . . . . . . . . . . . . . . . . 41 154 4.2.3.3. Feedback Control . . . . . . . . . . . . . . . . 42 155 4.2.4. Security Considerations . . . . . . . . . . . . . . . 42 156 4.3. BAS Future . . . . . . . . . . . . . . . . . . . . . . . 42 157 4.4. BAS Asks . . . . . . . . . . . . . . . . . . . . . . . . 43 158 5. Wireless for Industrial . . . . . . . . . . . . . . . . . . . 43 159 5.1. Use Case Description . . . . . . . . . . . . . . . . . . 43 160 5.1.1. Network Convergence using 6TiSCH . . . . . . . . . . 44 161 5.1.2. Common Protocol Development for 6TiSCH . . . . . . . 44 162 5.2. Wireless Industrial Today . . . . . . . . . . . . . . . . 45 163 5.3. Wireless Industrial Future . . . . . . . . . . . . . . . 45 164 5.3.1. Unified Wireless Network and Management . . . . . . . 45 165 5.3.1.1. PCE and 6TiSCH ARQ Retries . . . . . . . . . . . 47 166 5.3.2. Schedule Management by a PCE . . . . . . . . . . . . 48 167 5.3.2.1. PCE Commands and 6TiSCH CoAP Requests . . . . . . 48 168 5.3.2.2. 6TiSCH IP Interface . . . . . . . . . . . . . . . 49 169 5.3.3. 6TiSCH Security Considerations . . . . . . . . . . . 50 170 5.4. Wireless Industrial Asks . . . . . . . . . . . . . . . . 50 171 6. Cellular Radio . . . . . . . . . . . . . . . . . . . . . . . 50 172 6.1. Use Case Description . . . . . . . . . . . . . . . . . . 50 173 6.1.1. Network Architecture . . . . . . . . . . . . . . . . 50 174 6.1.2. Delay Constraints . . . . . . . . . . . . . . . . . . 51 175 6.1.3. Time Synchronization Constraints . . . . . . . . . . 53 176 6.1.4. Transport Loss Constraints . . . . . . . . . . . . . 55 177 6.1.5. Security Considerations . . . . . . . . . . . . . . . 55 178 6.2. Cellular Radio Networks Today . . . . . . . . . . . . . . 56 179 6.2.1. Fronthaul . . . . . . . . . . . . . . . . . . . . . . 56 180 6.2.2. Midhaul and Backhaul . . . . . . . . . . . . . . . . 56 181 6.3. Cellular Radio Networks Future . . . . . . . . . . . . . 57 182 6.4. Cellular Radio Networks Asks . . . . . . . . . . . . . . 59 183 7. Industrial M2M . . . . . . . . . . . . . . . . . . . . . . . 59 184 7.1. Use Case Description . . . . . . . . . . . . . . . . . . 59 185 7.2. Industrial M2M Communication Today . . . . . . . . . . . 60 186 7.2.1. Transport Parameters . . . . . . . . . . . . . . . . 61 187 7.2.2. Stream Creation and Destruction . . . . . . . . . . . 62 188 7.3. Industrial M2M Future . . . . . . . . . . . . . . . . . . 62 189 7.4. Industrial M2M Asks . . . . . . . . . . . . . . . . . . . 62 191 8. Mining Industry . . . . . . . . . . . . . . . . . . . . . . . 63 192 8.1. Use Case Description . . . . . . . . . . . . . . . . . . 63 193 8.2. Mining Industry Today . . . . . . . . . . . . . . . . . . 63 194 8.3. Mining Industry Future . . . . . . . . . . . . . . . . . 64 195 8.4. Mining Industry Asks . . . . . . . . . . . . . . . . . . 65 196 9. Private Blockchain . . . . . . . . . . . . . . . . . . . . . 65 197 9.1. Use Case Description . . . . . . . . . . . . . . . . . . 65 198 9.1.1. Blockchain Operation . . . . . . . . . . . . . . . . 65 199 9.1.2. Blockchain Network Architecture . . . . . . . . . . . 66 200 9.1.3. Security Considerations . . . . . . . . . . . . . . . 66 201 9.2. Private Blockchain Today . . . . . . . . . . . . . . . . 66 202 9.3. Private Blockchain Future . . . . . . . . . . . . . . . . 67 203 9.4. Private Blockchain Asks . . . . . . . . . . . . . . . . . 67 204 10. Network Slicing . . . . . . . . . . . . . . . . . . . . . . . 67 205 10.1. Use Case Description . . . . . . . . . . . . . . . . . . 67 206 10.2. Network Slicing Use Cases . . . . . . . . . . . . . . . 68 207 10.2.1. Enhanced Mobile Broadband (eMBB) . . . . . . . . . . 68 208 10.2.2. Ultra-Reliable and Low Latency Communications 209 (URLLC) . . . . . . . . . . . . . . . . . . . . . . 68 210 10.2.3. massive Machine Type Communications (mMTC) . . . . . 68 211 10.3. Using DetNet in Network Slicing . . . . . . . . . . . . 68 212 10.4. Network Slicing Today and Future . . . . . . . . . . . . 69 213 10.5. Network Slicing Asks . . . . . . . . . . . . . . . . . . 69 214 11. Use Case Common Themes . . . . . . . . . . . . . . . . . . . 69 215 11.1. Unified, standards-based network . . . . . . . . . . . . 69 216 11.1.1. Extensions to Ethernet . . . . . . . . . . . . . . . 69 217 11.1.2. Centrally Administered . . . . . . . . . . . . . . . 69 218 11.1.3. Standardized Data Flow Information Models . . . . . 70 219 11.1.4. L2 and L3 Integration . . . . . . . . . . . . . . . 70 220 11.1.5. Guaranteed End-to-End Delivery . . . . . . . . . . . 70 221 11.1.6. Replacement for Multiple Proprietary Deterministic 222 Networks . . . . . . . . . . . . . . . . . . . . . . 70 223 11.1.7. Mix of Deterministic and Best-Effort Traffic . . . . 70 224 11.1.8. Unused Reserved BW to be Available to Best Effort 225 Traffic . . . . . . . . . . . . . . . . . . . . . . 70 226 11.1.9. Lower Cost, Multi-Vendor Solutions . . . . . . . . . 71 227 11.2. Scalable Size . . . . . . . . . . . . . . . . . . . . . 71 228 11.3. Scalable Timing Parameters and Accuracy . . . . . . . . 71 229 11.3.1. Bounded Latency . . . . . . . . . . . . . . . . . . 71 230 11.3.2. Low Latency . . . . . . . . . . . . . . . . . . . . 71 231 11.3.3. Symmetrical Path Delays . . . . . . . . . . . . . . 72 232 11.4. High Reliability and Availability . . . . . . . . . . . 72 233 11.5. Security . . . . . . . . . . . . . . . . . . . . . . . . 72 234 11.6. Deterministic Flows . . . . . . . . . . . . . . . . . . 72 235 12. Use Cases Explicitly Out of Scope for DetNet . . . . . . . . 72 236 12.1. DetNet Scope Limitations . . . . . . . . . . . . . . . . 73 237 12.2. Internet-based Applications . . . . . . . . . . . . . . 73 238 12.2.1. Use Case Description . . . . . . . . . . . . . . . . 73 239 12.2.1.1. Media Content Delivery . . . . . . . . . . . . . 74 240 12.2.1.2. Online Gaming . . . . . . . . . . . . . . . . . 74 241 12.2.1.3. Virtual Reality . . . . . . . . . . . . . . . . 74 242 12.2.2. Internet-Based Applications Today . . . . . . . . . 74 243 12.2.3. Internet-Based Applications Future . . . . . . . . . 74 244 12.2.4. Internet-Based Applications Asks . . . . . . . . . . 74 245 12.3. Pro Audio and Video - Digital Rights Management (DRM) . 75 246 12.4. Pro Audio and Video - Link Aggregation . . . . . . . . . 75 247 13. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 76 248 13.1. Pro Audio . . . . . . . . . . . . . . . . . . . . . . . 76 249 13.2. Utility Telecom . . . . . . . . . . . . . . . . . . . . 76 250 13.3. Building Automation Systems . . . . . . . . . . . . . . 76 251 13.4. Wireless for Industrial . . . . . . . . . . . . . . . . 76 252 13.5. Cellular Radio . . . . . . . . . . . . . . . . . . . . . 77 253 13.6. Industrial M2M . . . . . . . . . . . . . . . . . . . . . 77 254 13.7. Internet Applications and CoMP . . . . . . . . . . . . . 77 255 13.8. Electrical Utilities . . . . . . . . . . . . . . . . . . 77 256 13.9. Network Slicing . . . . . . . . . . . . . . . . . . . . 77 257 13.10. Mining . . . . . . . . . . . . . . . . . . . . . . . . . 77 258 13.11. Private Blockchain . . . . . . . . . . . . . . . . . . . 77 259 14. Informative References . . . . . . . . . . . . . . . . . . . 77 260 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 87 262 1. Introduction 264 This draft presents use cases from diverse industries which have in 265 common a need for deterministic streams, but which also differ 266 notably in their network topologies and specific desired behavior. 267 Together, they provide broad industry context for DetNet and a 268 yardstick against which proposed DetNet designs can be measured (to 269 what extent does a proposed design satisfy these various use cases?) 271 For DetNet, use cases explicitly do not define requirements; The 272 DetNet WG will consider the use cases, decide which elements are in 273 scope for DetNet, and the results will be incorporated into future 274 drafts. Similarly, the DetNet use case draft explicitly does not 275 suggest any specific design, architecture or protocols, which will be 276 topics of future drafts. 278 We present for each use case the answers to the following questions: 280 o What is the use case? 282 o How is it addressed today? 284 o How would you like it to be addressed in the future? 286 o What do you want the IETF to deliver? 287 The level of detail in each use case should be sufficient to express 288 the relevant elements of the use case, but not more. 290 At the end we consider the use cases collectively, and examine the 291 most significant goals they have in common. 293 2. Pro Audio and Video 295 2.1. Use Case Description 297 The professional audio and video industry ("ProAV") includes: 299 o Music and film content creation 301 o Broadcast 303 o Cinema 305 o Live sound 307 o Public address, media and emergency systems at large venues 308 (airports, stadiums, churches, theme parks). 310 These industries have already transitioned audio and video signals 311 from analog to digital. However, the digital interconnect systems 312 remain primarily point-to-point with a single (or small number of) 313 signals per link, interconnected with purpose-built hardware. 315 These industries are now transitioning to packet-based infrastructure 316 to reduce cost, increase routing flexibility, and integrate with 317 existing IT infrastructure. 319 Today ProAV applications have no way to establish deterministic 320 streams from a standards-based Layer 3 (IP) interface, which is a 321 fundamental limitation to the use cases described here. Today 322 deterministic streams can be created within standards-based layer 2 323 LANs (e.g. using IEEE 802.1 AVB) however these are not routable via 324 IP and thus are not effective for distribution over wider areas (for 325 example broadcast events that span wide geographical areas). 327 It would be highly desirable if such streams could be routed over the 328 open Internet, however solutions with more limited scope (e.g. 329 enterprise networks) would still provide a substantial improvement. 331 The following sections describe specific ProAV use cases. 333 2.1.1. Uninterrupted Stream Playback 335 Transmitting audio and video streams for live playback is unlike 336 common file transfer because uninterrupted stream playback in the 337 presence of network errors cannot be achieved by re-trying the 338 transmission; by the time the missing or corrupt packet has been 339 identified it is too late to execute a re-try operation. Buffering 340 can be used to provide enough delay to allow time for one or more 341 retries, however this is not an effective solution in applications 342 where large delays (latencies) are not acceptable (as discussed 343 below). 345 Streams with guaranteed bandwidth can eliminate congestion on the 346 network as a cause of transmission errors that would lead to playback 347 interruption. Use of redundant paths can further mitigate 348 transmission errors to provide greater stream reliability. 350 2.1.2. Synchronized Stream Playback 352 Latency in this context is the time between when a signal is 353 initially sent over a stream and when it is received. A common 354 example in ProAV is time-synchronizing audio and video when they take 355 separate paths through the playback system. In this case the latency 356 of both the audio and video streams must be bounded and consistent if 357 the sound is to remain matched to the movement in the video. A 358 common tolerance for audio/video sync is one NTSC video frame (about 359 33ms) and to maintain the audience perception of correct lip sync the 360 latency needs to be consistent within some reasonable tolerance, for 361 example 10%. 363 A common architecture for synchronizing multiple streams that have 364 different paths through the network (and thus potentially different 365 latencies) is to enable measurement of the latency of each path, and 366 have the data sinks (for example speakers) delay (buffer) all packets 367 on all but the slowest path. Each packet of each stream is assigned 368 a presentation time which is based on the longest required delay. 369 This implies that all sinks must maintain a common time reference of 370 sufficient accuracy, which can be achieved by any of various 371 techniques. 373 This type of architecture is commonly implemented using a central 374 controller that determines path delays and arbitrates buffering 375 delays. 377 2.1.3. Sound Reinforcement 379 Consider the latency (delay) from when a person speaks into a 380 microphone to when their voice emerges from the speaker. If this 381 delay is longer than about 10-15 milliseconds it is noticeable and 382 can make a sound reinforcement system unusable (see slide 6 of 383 [SRP_LATENCY]). (If you have ever tried to speak in the presence of 384 a delayed echo of your voice you may know this experience). 386 Note that the 15ms latency bound includes all parts of the signal 387 path, not just the network, so the network latency must be 388 significantly less than 15ms. 390 In some cases local performers must perform in synchrony with a 391 remote broadcast. In such cases the latencies of the broadcast 392 stream and the local performer must be adjusted to match each other, 393 with a worst case of one video frame (33ms for NTSC video). 395 In cases where audio phase is a consideration, for example beam- 396 forming using multiple speakers, latency requirements can be in the 397 10 microsecond range (1 audio sample at 96kHz). 399 2.1.4. Deterministic Time to Establish Streaming 401 Note: The WG has decided that guidelines for deterministic time to 402 establish stream startup is not within scope of DetNet. If bounded 403 timing of establishing or re-establish streams is required in a given 404 use case, it is up to the application/system to achieve this. (The 405 supporting text from this section has been removed as of draft 12). 407 2.1.5. Secure Transmission 409 2.1.5.1. Safety 411 Professional audio systems can include amplifiers that are capable of 412 generating hundreds or thousands of watts of audio power which if 413 used incorrectly can cause hearing damage to those in the vicinity. 414 Apart from the usual care required by the systems operators to 415 prevent such incidents, the network traffic that controls these 416 devices must be secured (as with any sensitive application traffic). 418 2.2. Pro Audio Today 420 Some proprietary systems have been created which enable deterministic 421 streams at Layer 3 however they are "engineered networks" which 422 require careful configuration to operate, often require that the 423 system be over-provisioned, and it is implied that all devices on the 424 network voluntarily play by the rules of that network. To enable 425 these industries to successfully transition to an interoperable 426 multi-vendor packet-based infrastructure requires effective open 427 standards, and we believe that establishing relevant IETF standards 428 is a crucial factor. 430 2.3. Pro Audio Future 432 2.3.1. Layer 3 Interconnecting Layer 2 Islands 434 It would be valuable to enable IP to connect multiple Layer 2 LANs. 436 As an example, ESPN recently constructed a state-of-the-art 194,000 437 sq ft, $125 million broadcast studio called DC2. The DC2 network is 438 capable of handling 46 Tbps of throughput with 60,000 simultaneous 439 signals. Inside the facility are 1,100 miles of fiber feeding four 440 audio control rooms (see [ESPN_DC2] ). 442 In designing DC2 they replaced as much point-to-point technology as 443 they could with packet-based technology. They constructed seven 444 individual studios using layer 2 LANS (using IEEE 802.1 AVB) that 445 were entirely effective at routing audio within the LANs. However to 446 interconnect these layer 2 LAN islands together they ended up using 447 dedicated paths in a custom SDN (Software Defined Networking) router 448 because there is no standards-based routing solution available. 450 2.3.2. High Reliability Stream Paths 452 On-air and other live media streams are often backed up with 453 redundant links that seamlessly act to deliver the content when the 454 primary link fails for any reason. In point-to-point systems this is 455 provided by an additional point-to-point link; the analogous 456 requirement in a packet-based system is to provide an alternate path 457 through the network such that no individual link can bring down the 458 system. 460 2.3.3. Integration of Reserved Streams into IT Networks 462 A commonly cited goal of moving to a packet based media 463 infrastructure is that costs can be reduced by using off the shelf, 464 commodity network hardware. In addition, economy of scale can be 465 realized by combining media infrastructure with IT infrastructure. 466 In keeping with these goals, stream reservation technology should be 467 compatible with existing protocols, and not compromise use of the 468 network for best effort (non-time-sensitive) traffic. 470 2.3.4. Use of Unused Reservations by Best-Effort Traffic 472 In cases where stream bandwidth is reserved but not currently used 473 (or is under-utilized) that bandwidth must be available to best- 474 effort (i.e. non-time-sensitive) traffic. For example a single 475 stream may be nailed up (reserved) for specific media content that 476 needs to be presented at different times of the day, ensuring timely 477 delivery of that content, yet in between those times the full 478 bandwidth of the network can be utilized for best-effort tasks such 479 as file transfers. 481 This also addresses a concern of IT network administrators that are 482 considering adding reserved bandwidth traffic to their networks that 483 ("users will reserve large quantities of bandwidth and then never un- 484 reserve it even though they are not using it, and soon the network 485 will have no bandwidth left"). 487 2.3.5. Traffic Segregation 489 Note: It is still under WG discussion whether this topic will be 490 addressed by DetNet. 492 Sink devices may be low cost devices with limited processing power. 493 In order to not overwhelm the CPUs in these devices it is important 494 to limit the amount of traffic that these devices must process. 496 As an example, consider the use of individual seat speakers in a 497 cinema. These speakers are typically required to be cost reduced 498 since the quantities in a single theater can reach hundreds of seats. 499 Discovery protocols alone in a one thousand seat theater can generate 500 enough broadcast traffic to overwhelm a low powered CPU. Thus an 501 installation like this will benefit greatly from some type of traffic 502 segregation that can define groups of seats to reduce traffic within 503 each group. All seats in the theater must still be able to 504 communicate with a central controller. 506 There are many techniques that can be used to support this 507 requirement including (but not limited to) the following examples. 509 2.3.5.1. Packet Forwarding Rules, VLANs and Subnets 511 Packet forwarding rules can be used to eliminate some extraneous 512 streaming traffic from reaching potentially low powered sink devices, 513 however there may be other types of broadcast traffic that should be 514 eliminated using other means for example VLANs or IP subnets. 516 2.3.5.2. Multicast Addressing (IPv4 and IPv6) 518 Multicast addressing is commonly used to keep bandwidth utilization 519 of shared links to a minimum. 521 Because of the MAC Address forwarding nature of Layer 2 bridges it is 522 important that a multicast MAC address is only associated with one 523 stream. This will prevent reservations from forwarding packets from 524 one stream down a path that has no interested sinks simply because 525 there is another stream on that same path that shares the same 526 multicast MAC address. 528 Since each multicast MAC Address can represent 32 different IPv4 529 multicast addresses there must be a process put in place to make sure 530 this does not occur. Requiring use of IPv6 address can achieve this, 531 however due to their continued prevalence, solutions that are 532 effective for IPv4 installations are also required. 534 2.3.6. Latency Optimization by a Central Controller 536 A central network controller might also perform optimizations based 537 on the individual path delays, for example sinks that are closer to 538 the source can inform the controller that they can accept greater 539 latency since they will be buffering packets to match presentation 540 times of farther away sinks. The controller might then move a stream 541 reservation on a short path to a longer path in order to free up 542 bandwidth for other critical streams on that short path. See slides 543 3-5 of [SRP_LATENCY]. 545 Additional optimization can be achieved in cases where sinks have 546 differing latency requirements, for example in a live outdoor concert 547 the speaker sinks have stricter latency requirements than the 548 recording hardware sinks. See slide 7 of [SRP_LATENCY]. 550 2.3.7. Reduced Device Cost Due To Reduced Buffer Memory 552 Device cost can be reduced in a system with guaranteed reservations 553 with a small bounded latency due to the reduced requirements for 554 buffering (i.e. memory) on sink devices. For example, a theme park 555 might broadcast a live event across the globe via a layer 3 protocol; 556 in such cases the size of the buffers required is proportional to the 557 latency bounds and jitter caused by delivery, which depends on the 558 worst case segment of the end-to-end network path. For example on 559 todays open internet the latency is typically unacceptable for audio 560 and video streaming without many seconds of buffering. In such 561 scenarios a single gateway device at the local network that receives 562 the feed from the remote site would provide the expensive buffering 563 required to mask the latency and jitter issues associated with long 564 distance delivery. Sink devices in the local location would have no 565 additional buffering requirements, and thus no additional costs, 566 beyond those required for delivery of local content. The sink device 567 would be receiving the identical packets as those sent by the source 568 and would be unaware that there were any latency or jitter issues 569 along the path. 571 2.4. Pro Audio Asks 573 o Layer 3 routing on top of AVB (and/or other high QoS networks) 575 o Content delivery with bounded, lowest possible latency 577 o IntServ and DiffServ integration with AVB (where practical) 579 o Single network for A/V and IT traffic 581 o Standards-based, interoperable, multi-vendor 583 o IT department friendly 585 o Enterprise-wide networks (e.g. size of San Francisco but not the 586 whole Internet (yet...)) 588 3. Electrical Utilities 590 3.1. Use Case Description 592 Many systems that an electrical utility deploys today rely on high 593 availability and deterministic behavior of the underlying networks. 594 Here we present use cases in Transmission, Generation and 595 Distribution, including key timing and reliability metrics. We also 596 discuss security issues and industry trends which affect the 597 architecture of next generation utility networks 599 3.1.1. Transmission Use Cases 601 3.1.1.1. Protection 603 Protection means not only the protection of human operators but also 604 the protection of the electrical equipment and the preservation of 605 the stability and frequency of the grid. If a fault occurs in the 606 transmission or distribution of electricity then severe damage can 607 occur to human operators, electrical equipment and the grid itself, 608 leading to blackouts. 610 Communication links in conjunction with protection relays are used to 611 selectively isolate faults on high voltage lines, transformers, 612 reactors and other important electrical equipment. The role of the 613 teleprotection system is to selectively disconnect a faulty part by 614 transferring command signals within the shortest possible time. 616 3.1.1.1.1. Key Criteria 618 The key criteria for measuring teleprotection performance are command 619 transmission time, dependability and security. These criteria are 620 defined by the IEC standard 60834 as follows: 622 o Transmission time (Speed): The time between the moment where state 623 changes at the transmitter input and the moment of the 624 corresponding change at the receiver output, including propagation 625 delay. Overall operating time for a teleprotection system 626 includes the time for initiating the command at the transmitting 627 end, the propagation delay over the network (including equipments) 628 and the selection and decision time at the receiving end, 629 including any additional delay due to a noisy environment. 631 o Dependability: The ability to issue and receive valid commands in 632 the presence of interference and/or noise, by minimizing the 633 probability of missing command (PMC). Dependability targets are 634 typically set for a specific bit error rate (BER) level. 636 o Security: The ability to prevent false tripping due to a noisy 637 environment, by minimizing the probability of unwanted commands 638 (PUC). Security targets are also set for a specific bit error 639 rate (BER) level. 641 Additional elements of the the teleprotection system that impact its 642 performance include: 644 o Network bandwidth 646 o Failure recovery capacity (aka resiliency) 648 3.1.1.1.2. Fault Detection and Clearance Timing 650 Most power line equipment can tolerate short circuits or faults for 651 up to approximately five power cycles before sustaining irreversible 652 damage or affecting other segments in the network. This translates 653 to total fault clearance time of 100ms. As a safety precaution, 654 however, actual operation time of protection systems is limited to 655 70- 80 percent of this period, including fault recognition time, 656 command transmission time and line breaker switching time. 658 Some system components, such as large electromechanical switches, 659 require particularly long time to operate and take up the majority of 660 the total clearance time, leaving only a 10ms window for the 661 telecommunications part of the protection scheme, independent of the 662 distance to travel. Given the sensitivity of the issue, new networks 663 impose requirements that are even more stringent: IEC standard 61850 664 limits the transfer time for protection messages to 1/4 - 1/2 cycle 665 or 4 - 8ms (for 60Hz lines) for the most critical messages. 667 3.1.1.1.3. Symmetric Channel Delay 669 Note: It is currently under WG discussion whether symmetric path 670 delays are to be guaranteed by DetNet. 672 Teleprotection channels which are differential must be synchronous, 673 which means that any delays on the transmit and receive paths must 674 match each other. Teleprotection systems ideally support zero 675 asymmetric delay; typical legacy relays can tolerate delay 676 discrepancies of up to 750us. 678 Some tools available for lowering delay variation below this 679 threshold are: 681 o For legacy systems using Time Division Multiplexing (TDM), jitter 682 buffers at the multiplexers on each end of the line can be used to 683 offset delay variation by queuing sent and received packets. The 684 length of the queues must balance the need to regulate the rate of 685 transmission with the need to limit overall delay, as larger 686 buffers result in increased latency. 688 o For jitter-prone IP packet networks, traffic management tools can 689 ensure that the teleprotection signals receive the highest 690 transmission priority to minimize jitter. 692 o Standard packet-based synchronization technologies, such as 693 1588-2008 Precision Time Protocol (PTP) and Synchronous Ethernet 694 (Sync-E), can help keep networks stable by maintaining a highly 695 accurate clock source on the various network devices. 697 3.1.1.1.4. Teleprotection Network Requirements (IEC 61850) 699 The following table captures the main network metrics as based on the 700 IEC 61850 standard. 702 +-----------------------------+-------------------------------------+ 703 | Teleprotection Requirement | Attribute | 704 +-----------------------------+-------------------------------------+ 705 | One way maximum delay | 4-10 ms | 706 | Asymetric delay required | Yes | 707 | Maximum jitter | less than 250 us (750 us for legacy | 708 | | IED) | 709 | Topology | Point to point, point to Multi- | 710 | | point | 711 | Availability | 99.9999 | 712 | precise timing required | Yes | 713 | Recovery time on node | less than 50ms - hitless | 714 | failure | | 715 | performance management | Yes, Mandatory | 716 | Redundancy | Yes | 717 | Packet loss | 0.1% to 1% | 718 +-----------------------------+-------------------------------------+ 720 Table 1: Teleprotection network requirements 722 3.1.1.1.5. Inter-Trip Protection scheme 724 "Inter-tripping" is the signal-controlled tripping of a circuit 725 breaker to complete the isolation of a circuit or piece of apparatus 726 in concert with the tripping of other circuit breakers. 728 +--------------------------------+----------------------------------+ 729 | Inter-Trip protection | Attribute | 730 | Requirement | | 731 +--------------------------------+----------------------------------+ 732 | One way maximum delay | 5 ms | 733 | Asymetric delay required | No | 734 | Maximum jitter | Not critical | 735 | Topology | Point to point, point to Multi- | 736 | | point | 737 | Bandwidth | 64 Kbps | 738 | Availability | 99.9999 | 739 | precise timing required | Yes | 740 | Recovery time on node failure | less than 50ms - hitless | 741 | performance management | Yes, Mandatory | 742 | Redundancy | Yes | 743 | Packet loss | 0.1% | 744 +--------------------------------+----------------------------------+ 746 Table 2: Inter-Trip protection network requirements 748 3.1.1.1.6. Current Differential Protection Scheme 750 Current differential protection is commonly used for line protection, 751 and is typical for protecting parallel circuits. At both end of the 752 lines the current is measured by the differential relays, and both 753 relays will trip the circuit breaker if the current going into the 754 line does not equal the current going out of the line. This type of 755 protection scheme assumes some form of communications being present 756 between the relays at both end of the line, to allow both relays to 757 compare measured current values. Line differential protection 758 schemes assume a very low telecommunications delay between both 759 relays, often as low as 5ms. Moreover, as those systems are often 760 not time-synchronized, they also assume symmetric telecommunications 761 paths with constant delay, which allows comparing current measurement 762 values taken at the exact same time. 764 +----------------------------------+--------------------------------+ 765 | Current Differential protection | Attribute | 766 | Requirement | | 767 +----------------------------------+--------------------------------+ 768 | One way maximum delay | 5 ms | 769 | Asymetric delay Required | Yes | 770 | Maximum jitter | less than 250 us (750us for | 771 | | legacy IED) | 772 | Topology | Point to point, point to | 773 | | Multi-point | 774 | Bandwidth | 64 Kbps | 775 | Availability | 99.9999 | 776 | precise timing required | Yes | 777 | Recovery time on node failure | less than 50ms - hitless | 778 | performance management | Yes, Mandatory | 779 | Redundancy | Yes | 780 | Packet loss | 0.1% | 781 +----------------------------------+--------------------------------+ 783 Table 3: Current Differential Protection metrics 785 3.1.1.1.7. Distance Protection Scheme 787 Distance (Impedance Relay) protection scheme is based on voltage and 788 current measurements. The network metrics are similar (but not 789 identical to) Current Differential protection. 791 +-------------------------------+-----------------------------------+ 792 | Distance protection | Attribute | 793 | Requirement | | 794 +-------------------------------+-----------------------------------+ 795 | One way maximum delay | 5 ms | 796 | Asymetric delay Required | No | 797 | Maximum jitter | Not critical | 798 | Topology | Point to point, point to Multi- | 799 | | point | 800 | Bandwidth | 64 Kbps | 801 | Availability | 99.9999 | 802 | precise timing required | Yes | 803 | Recovery time on node failure | less than 50ms - hitless | 804 | performance management | Yes, Mandatory | 805 | Redundancy | Yes | 806 | Packet loss | 0.1% | 807 +-------------------------------+-----------------------------------+ 809 Table 4: Distance Protection requirements 811 3.1.1.1.8. Inter-Substation Protection Signaling 813 This use case describes the exchange of Sampled Value and/or GOOSE 814 (Generic Object Oriented Substation Events) message between 815 Intelligent Electronic Devices (IED) in two substations for 816 protection and tripping coordination. The two IEDs are in a master- 817 slave mode. 819 The Current Transformer or Voltage Transformer (CT/VT) in one 820 substation sends the sampled analog voltage or current value to the 821 Merging Unit (MU) over hard wire. The MU sends the time-synchronized 822 61850-9-2 sampled values to the slave IED. The slave IED forwards 823 the information to the Master IED in the other substation. The 824 master IED makes the determination (for example based on sampled 825 value differentials) to send a trip command to the originating IED. 826 Once the slave IED/Relay receives the GOOSE trip for breaker 827 tripping, it opens the breaker. It then sends a confirmation message 828 back to the master. All data exchanges between IEDs are either 829 through Sampled Value and/or GOOSE messages. 831 +----------------------------------+--------------------------------+ 832 | Inter-Substation protection | Attribute | 833 | Requirement | | 834 +----------------------------------+--------------------------------+ 835 | One way maximum delay | 5 ms | 836 | Asymetric delay Required | No | 837 | Maximum jitter | Not critical | 838 | Topology | Point to point, point to | 839 | | Multi-point | 840 | Bandwidth | 64 Kbps | 841 | Availability | 99.9999 | 842 | precise timing required | Yes | 843 | Recovery time on node failure | less than 50ms - hitless | 844 | performance management | Yes, Mandatory | 845 | Redundancy | Yes | 846 | Packet loss | 1% | 847 +----------------------------------+--------------------------------+ 849 Table 5: Inter-Substation Protection requirements 851 3.1.1.2. Intra-Substation Process Bus Communications 853 This use case describes the data flow from the CT/VT to the IEDs in 854 the substation via the MU. The CT/VT in the substation send the 855 analog voltage or current values to the MU over hard wire. The MU 856 converts the analog values into digital format (typically time- 857 synchronized Sampled Values as specified by IEC 61850-9-2) and sends 858 them to the IEDs in the substation. The GPS Master Clock can send 859 1PPS or IRIG-B format to the MU through a serial port or IEEE 1588 860 protocol via a network. Process bus communication using 61850 861 simplifies connectivity within the substation and removes the 862 requirement for multiple serial connections and removes the slow 863 serial bus architectures that are typically used. This also ensures 864 increased flexibility and increased speed with the use of multicast 865 messaging between multiple devices. 867 +----------------------------------+--------------------------------+ 868 | Intra-Substation protection | Attribute | 869 | Requirement | | 870 +----------------------------------+--------------------------------+ 871 | One way maximum delay | 5 ms | 872 | Asymetric delay Required | No | 873 | Maximum jitter | Not critical | 874 | Topology | Point to point, point to | 875 | | Multi-point | 876 | Bandwidth | 64 Kbps | 877 | Availability | 99.9999 | 878 | precise timing required | Yes | 879 | Recovery time on Node failure | less than 50ms - hitless | 880 | performance management | Yes, Mandatory | 881 | Redundancy | Yes - No | 882 | Packet loss | 0.1% | 883 +----------------------------------+--------------------------------+ 885 Table 6: Intra-Substation Protection requirements 887 3.1.1.3. Wide Area Monitoring and Control Systems 889 The application of synchrophasor measurement data from Phasor 890 Measurement Units (PMU) to Wide Area Monitoring and Control Systems 891 promises to provide important new capabilities for improving system 892 stability. Access to PMU data enables more timely situational 893 awareness over larger portions of the grid than what has been 894 possible historically with normal SCADA (Supervisory Control and Data 895 Acquisition) data. Handling the volume and real-time nature of 896 synchrophasor data presents unique challenges for existing 897 application architectures. Wide Area management System (WAMS) makes 898 it possible for the condition of the bulk power system to be observed 899 and understood in real-time so that protective, preventative, or 900 corrective action can be taken. Because of the very high sampling 901 rate of measurements and the strict requirement for time 902 synchronization of the samples, WAMS has stringent telecommunications 903 requirements in an IP network that are captured in the following 904 table: 906 +----------------------+--------------------------------------------+ 907 | WAMS Requirement | Attribute | 908 +----------------------+--------------------------------------------+ 909 | One way maximum | 50 ms | 910 | delay | | 911 | Asymetric delay | No | 912 | Required | | 913 | Maximum jitter | Not critical | 914 | Topology | Point to point, point to Multi-point, | 915 | | Multi-point to Multi-point | 916 | Bandwidth | 100 Kbps | 917 | Availability | 99.9999 | 918 | precise timing | Yes | 919 | required | | 920 | Recovery time on | less than 50ms - hitless | 921 | Node failure | | 922 | performance | Yes, Mandatory | 923 | management | | 924 | Redundancy | Yes | 925 | Packet loss | 1% | 926 | Consecutive Packet | At least 1 packet per application cycle | 927 | Loss | must be received. | 928 +----------------------+--------------------------------------------+ 930 Table 7: WAMS Special Communication Requirements 932 3.1.1.4. IEC 61850 WAN engineering guidelines requirement 933 classification 935 The IEC (International Electrotechnical Commission) has recently 936 published a Technical Report which offers guidelines on how to define 937 and deploy Wide Area Networks for the interconnections of electric 938 substations, generation plants and SCADA operation centers. The IEC 939 61850-90-12 is providing a classification of WAN communication 940 requirements into 4 classes. Table 8 summarizes these requirements: 942 +----------------+------------+------------+------------+-----------+ 943 | WAN | Class WA | Class WB | Class WC | Class WD | 944 | Requirement | | | | | 945 +----------------+------------+------------+------------+-----------+ 946 | Application | EHV (Extra | HV (High | MV (Medium | General | 947 | field | High | Voltage) | Voltage) | purpose | 948 | | Voltage) | | | | 949 | Latency | 5 ms | 10 ms | 100 ms | > 100 ms | 950 | Jitter | 10 us | 100 us | 1 ms | 10 ms | 951 | Latency | 100 us | 1 ms | 10 ms | 100 ms | 952 | Asymetry | | | | | 953 | Time Accuracy | 1 us | 10 us | 100 us | 10 to 100 | 954 | | | | | ms | 955 | Bit Error rate | 10-7 to | 10-5 to | 10-3 | | 956 | | 10-6 | 10-4 | | | 957 | Unavailability | 10-7 to | 10-5 to | 10-3 | | 958 | | 10-6 | 10-4 | | | 959 | Recovery delay | Zero | 50 ms | 5 s | 50 s | 960 | Cyber security | extremely | High | Medium | Medium | 961 | | high | | | | 962 +----------------+------------+------------+------------+-----------+ 964 Table 8: 61850-90-12 Communication Requirements; Courtesy of IEC 966 3.1.2. Generation Use Case 968 Energy generation systems are complex infrastructures that require 969 control of both the generated power and the generation 970 infrastructure. 972 3.1.2.1. Control of the Generated Power 974 The electrical power generation frequency must be maintained within a 975 very narrow band. Deviations from the acceptable frequency range are 976 detected and the required signals are sent to the power plants for 977 frequency regulation. 979 Automatic Generation Control (AGC) is a system for adjusting the 980 power output of generators at different power plants, in response to 981 changes in the load. 983 +---------------------------------------------------+---------------+ 984 | FCAG (Frequency Control Automatic Generation) | Attribute | 985 | Requirement | | 986 +---------------------------------------------------+---------------+ 987 | One way maximum delay | 500 ms | 988 | Asymetric delay Required | No | 989 | Maximum jitter | Not critical | 990 | Topology | Point to | 991 | | point | 992 | Bandwidth | 20 Kbps | 993 | Availability | 99.999 | 994 | precise timing required | Yes | 995 | Recovery time on Node failure | N/A | 996 | performance management | Yes, | 997 | | Mandatory | 998 | Redundancy | Yes | 999 | Packet loss | 1% | 1000 +---------------------------------------------------+---------------+ 1002 Table 9: FCAG Communication Requirements 1004 3.1.2.2. Control of the Generation Infrastructure 1006 The control of the generation infrastructure combines requirements 1007 from industrial automation systems and energy generation systems. In 1008 this section we present the use case of the control of the generation 1009 infrastructure of a wind turbine. 1011 | 1012 | 1013 | +-----------------+ 1014 | | +----+ | 1015 | | |WTRM| WGEN | 1016 WROT x==|===| | | 1017 | | +----+ WCNV| 1018 | |WNAC | 1019 | +---+---WYAW---+--+ 1020 | | | 1021 | | | +----+ 1022 |WTRF | |WMET| 1023 | | | | 1024 Wind Turbine | +--+-+ 1025 Controller | | 1026 WTUR | | | 1027 WREP | | | 1028 WSLG | | | 1029 WALG | WTOW | | 1031 Figure 1: Wind Turbine Control Network 1033 Figure 1 presents the subsystems that operate a wind turbine. These 1034 subsystems include 1036 o WROT (Rotor Control) 1038 o WNAC (Nacelle Control) (nacelle: housing containing the generator) 1040 o WTRM (Transmission Control) 1042 o WGEN (Generator) 1044 o WYAW (Yaw Controller) (of the tower head) 1046 o WCNV (In-Turbine Power Converter) 1048 o WMET (External Meteorological Station providing real time 1049 information to the controllers of the tower) 1051 Traffic characteristics relevant for the network planning and 1052 dimensioning process in a wind turbine scenario are listed below. 1053 The values in this section are based mainly on the relevant 1054 references [Ahm14] and [Spe09]. Each logical node (Figure 1) is a 1055 part of the metering network and produces analog measurements and 1056 status information which must comply with their respective data rate 1057 constraints. 1059 +-----------+--------+--------+-------------+---------+-------------+ 1060 | Subsystem | Sensor | Analog | Data Rate | Status | Data rate | 1061 | | Count | Sample | (bytes/sec) | Sample | (bytes/sec) | 1062 | | | Count | | Count | | 1063 +-----------+--------+--------+-------------+---------+-------------+ 1064 | WROT | 14 | 9 | 642 | 5 | 10 | 1065 | WTRM | 18 | 10 | 2828 | 8 | 16 | 1066 | WGEN | 14 | 12 | 73764 | 2 | 4 | 1067 | WCNV | 14 | 12 | 74060 | 2 | 4 | 1068 | WTRF | 12 | 5 | 73740 | 2 | 4 | 1069 | WNAC | 12 | 9 | 112 | 3 | 6 | 1070 | WYAW | 7 | 8 | 220 | 4 | 8 | 1071 | WTOW | 4 | 1 | 8 | 3 | 6 | 1072 | WMET | 7 | 7 | 228 | - | - | 1073 +-----------+--------+--------+-------------+---------+-------------+ 1075 Table 10: Wind Turbine Data Rate Constraints 1077 Quality of Service (QoS) constraints for different services are 1078 presented in Table 11. These constraints are defined by IEEE 1646 1079 standard [IEEE1646] and IEC 61400 standard [IEC61400]. 1081 +---------------------+---------+-------------+---------------------+ 1082 | Service | Latency | Reliability | Packet Loss Rate | 1083 +---------------------+---------+-------------+---------------------+ 1084 | Analogue measure | 16 ms | 99.99% | < 10-6 | 1085 | Status information | 16 ms | 99.99% | < 10-6 | 1086 | Protection traffic | 4 ms | 100.00% | < 10-9 | 1087 | Reporting and | 1 s | 99.99% | < 10-6 | 1088 | logging | | | | 1089 | Video surveillance | 1 s | 99.00% | No specific | 1090 | | | | requirement | 1091 | Internet connection | 60 min | 99.00% | No specific | 1092 | | | | requirement | 1093 | Control traffic | 16 ms | 100.00% | < 10-9 | 1094 | Data polling | 16 ms | 99.99% | < 10-6 | 1095 +---------------------+---------+-------------+---------------------+ 1097 Table 11: Wind Turbine Reliability and Latency Constraints 1099 3.1.2.2.1. Intra-Domain Network Considerations 1101 A wind turbine is composed of a large set of subsystems including 1102 sensors and actuators which require time-critical operation. The 1103 reliability and latency constraints of these different subsystems is 1104 shown in Table 11. These subsystems are connected to an intra-domain 1105 network which is used to monitor and control the operation of the 1106 turbine and connect it to the SCADA subsystems. The different 1107 components are interconnected using fiber optics, industrial buses, 1108 industrial Ethernet, EtherCat, or a combination of them. Industrial 1109 signaling and control protocols such as Modbus, Profibus, Profinet 1110 and EtherCat are used directly on top of the Layer 2 transport or 1111 encapsulated over TCP/IP. 1113 The Data collected from the sensors and condition monitoring systems 1114 is multiplexed onto fiber cables for transmission to the base of the 1115 tower, and to remote control centers. The turbine controller 1116 continuously monitors the condition of the wind turbine and collects 1117 statistics on its operation. This controller also manages a large 1118 number of switches, hydraulic pumps, valves, and motors within the 1119 wind turbine. 1121 There is usually a controller both at the bottom of the tower and in 1122 the nacelle. The communication between these two controllers usually 1123 takes place using fiber optics instead of copper links. Sometimes, a 1124 third controller is installed in the hub of the rotor and manages the 1125 pitch of the blades. That unit usually communicates with the nacelle 1126 unit using serial communications. 1128 3.1.2.2.2. Inter-Domain network considerations 1130 A remote control center belonging to a grid operator regulates the 1131 power output, enables remote actuation, and monitors the health of 1132 one or more wind parks in tandem. It connects to the local control 1133 center in a wind park over the Internet (Figure 2) via firewalls at 1134 both ends. The AS path between the local control center and the Wind 1135 Park typically involves several ISPs at different tiers. For 1136 example, a remote control center in Denmark can regulate a wind park 1137 in Greece over the normal public AS path between the two locations. 1139 The remote control center is part of the SCADA system, setting the 1140 desired power output to the wind park and reading back the result 1141 once the new power output level has been set. Traffic between the 1142 remote control center and the wind park typically consists of 1143 protocols like IEC 60870-5-104 [IEC-60870-5-104], OPC XML-DA 1144 [OPCXML], Modbus [MODBUS], and SNMP [RFC3411]. Currently, traffic 1145 flows between the wind farm and the remote control center are best 1146 effort. QoS requirements are not strict, so no SLAs or service 1147 provisioning mechanisms (e.g., VPN) are employed. In case of events 1148 like equipment failure, tolerance for alarm delay is on the order of 1149 minutes, due to redundant systems already in place. 1151 +--------------+ 1152 | | 1153 | | 1154 | Wind Park #1 +----+ 1155 | | | XXXXXX 1156 | | | X XXXXXXXX +----------------+ 1157 +--------------+ | XXXX X XXXXX | | 1158 +---+ XXX | Remote Control | 1159 XXX Internet +----+ Center | 1160 +----+X XXX | | 1161 +--------------+ | XXXXXXX XX | | 1162 | | | XX XXXXXXX +----------------+ 1163 | | | XXXXX 1164 | Wind Park #2 +----+ 1165 | | 1166 | | 1167 +--------------+ 1169 Figure 2: Wind Turbine Control via Internet 1171 We expect future use cases which require bounded latency, bounded 1172 jitter and extraordinary low packet loss for inter-domain traffic 1173 flows due to the softwarization and virtualization of core wind farm 1174 equipment (e.g. switches, firewalls and SCADA server components). 1175 These factors will create opportunities for service providers to 1176 install new services and dynamically manage them from remote 1177 locations. For example, to enable fail-over of a local SCADA server, 1178 a SCADA server in another wind farm site (under the administrative 1179 control of the same operator) could be utilized temporarily 1180 (Figure 3). In that case local traffic would be forwarded to the 1181 remote SCADA server and existing intra-domain QoS and timing 1182 parameters would have to be met for inter-domain traffic flows. 1184 +--------------+ 1185 | | 1186 | | 1187 | Wind Park #1 +----+ 1188 | | | XXXXXX 1189 | | | X XXXXXXXX +----------------+ 1190 +--------------+ | XXXX XXXXX | | 1191 +---+ Operator XXX | Remote Control | 1192 XXX Administered +----+ Center | 1193 +----+X WAN XXX | | 1194 +--------------+ | XXXXXXX XX | | 1195 | | | XX XXXXXXX +----------------+ 1196 | | | XXXXX 1197 | Wind Park #2 +----+ 1198 | | 1199 | | 1200 +--------------+ 1202 Figure 3: Wind Turbine Control via Operator Administered WAN 1204 3.1.3. Distribution use case 1206 3.1.3.1. Fault Location Isolation and Service Restoration (FLISR) 1208 Fault Location, Isolation, and Service Restoration (FLISR) refers to 1209 the ability to automatically locate the fault, isolate the fault, and 1210 restore service in the distribution network. This will likely be the 1211 first widespread application of distributed intelligence in the grid. 1213 Static power switch status (open/closed) in the network dictates the 1214 power flow to secondary substations. Reconfiguring the network in 1215 the event of a fault is typically done manually on site to energize/ 1216 de-energize alternate paths. Automating the operation of substation 1217 switchgear allows the flow of power to be altered automatically under 1218 fault conditions. 1220 FLISR can be managed centrally from a Distribution Management System 1221 (DMS) or executed locally through distributed control via intelligent 1222 switches and fault sensors. 1224 +----------------------+--------------------------------------------+ 1225 | FLISR Requirement | Attribute | 1226 +----------------------+--------------------------------------------+ 1227 | One way maximum | 80 ms | 1228 | delay | | 1229 | Asymetric delay | No | 1230 | Required | | 1231 | Maximum jitter | 40 ms | 1232 | Topology | Point to point, point to Multi-point, | 1233 | | Multi-point to Multi-point | 1234 | Bandwidth | 64 Kbps | 1235 | Availability | 99.9999 | 1236 | precise timing | Yes | 1237 | required | | 1238 | Recovery time on | Depends on customer impact | 1239 | Node failure | | 1240 | performance | Yes, Mandatory | 1241 | management | | 1242 | Redundancy | Yes | 1243 | Packet loss | 0.1% | 1244 +----------------------+--------------------------------------------+ 1246 Table 12: FLISR Communication Requirements 1248 3.2. Electrical Utilities Today 1250 Many utilities still rely on complex environments formed of multiple 1251 application-specific proprietary networks, including TDM networks. 1253 In this kind of environment there is no mixing of OT and IT 1254 applications on the same network, and information is siloed between 1255 operational areas. 1257 Specific calibration of the full chain is required, which is costly. 1259 This kind of environment prevents utility operations from realizing 1260 the operational efficiency benefits, visibility, and functional 1261 integration of operational information across grid applications and 1262 data networks. 1264 In addition, there are many security-related issues as discussed in 1265 the following section. 1267 3.2.1. Security Current Practices and Limitations 1269 Grid monitoring and control devices are already targets for cyber 1270 attacks, and legacy telecommunications protocols have many intrinsic 1271 network-related vulnerabilities. For example, DNP3, Modbus, 1272 PROFIBUS/PROFINET, and other protocols are designed around a common 1273 paradigm of request and respond. Each protocol is designed for a 1274 master device such as an HMI (Human Machine Interface) system to send 1275 commands to subordinate slave devices to retrieve data (reading 1276 inputs) or control (writing to outputs). Because many of these 1277 protocols lack authentication, encryption, or other basic security 1278 measures, they are prone to network-based attacks, allowing a 1279 malicious actor or attacker to utilize the request-and-respond system 1280 as a mechanism for command-and-control like functionality. Specific 1281 security concerns common to most industrial control, including 1282 utility telecommunication protocols include the following: 1284 o Network or transport errors (e.g. malformed packets or excessive 1285 latency) can cause protocol failure. 1287 o Protocol commands may be available that are capable of forcing 1288 slave devices into inoperable states, including powering-off 1289 devices, forcing them into a listen-only state, disabling 1290 alarming. 1292 o Protocol commands may be available that are capable of restarting 1293 communications and otherwise interrupting processes. 1295 o Protocol commands may be available that are capable of clearing, 1296 erasing, or resetting diagnostic information such as counters and 1297 diagnostic registers. 1299 o Protocol commands may be available that are capable of requesting 1300 sensitive information about the controllers, their configurations, 1301 or other need-to-know information. 1303 o Most protocols are application layer protocols transported over 1304 TCP; therefore it is easy to transport commands over non-standard 1305 ports or inject commands into authorized traffic flows. 1307 o Protocol commands may be available that are capable of 1308 broadcasting messages to many devices at once (i.e. a potential 1309 DoS). 1311 o Protocol commands may be available to query the device network to 1312 obtain defined points and their values (i.e. a configuration 1313 scan). 1315 o Protocol commands may be available that will list all available 1316 function codes (i.e. a function scan). 1318 These inherent vulnerabilities, along with increasing connectivity 1319 between IT an OT networks, make network-based attacks very feasible. 1321 Simple injection of malicious protocol commands provides control over 1322 the target process. Altering legitimate protocol traffic can also 1323 alter information about a process and disrupt the legitimate controls 1324 that are in place over that process. A man-in-the-middle attack 1325 could provide both control over a process and misrepresentation of 1326 data back to operator consoles. 1328 3.3. Electrical Utilities Future 1330 The business and technology trends that are sweeping the utility 1331 industry will drastically transform the utility business from the way 1332 it has been for many decades. At the core of many of these changes 1333 is a drive to modernize the electrical grid with an integrated 1334 telecommunications infrastructure. However, interoperability 1335 concerns, legacy networks, disparate tools, and stringent security 1336 requirements all add complexity to the grid transformation. Given 1337 the range and diversity of the requirements that should be addressed 1338 by the next generation telecommunications infrastructure, utilities 1339 need to adopt a holistic architectural approach to integrate the 1340 electrical grid with digital telecommunications across the entire 1341 power delivery chain. 1343 The key to modernizing grid telecommunications is to provide a 1344 common, adaptable, multi-service network infrastructure for the 1345 entire utility organization. Such a network serves as the platform 1346 for current capabilities while enabling future expansion of the 1347 network to accommodate new applications and services. 1349 To meet this diverse set of requirements, both today and in the 1350 future, the next generation utility telecommunnications network will 1351 be based on open-standards-based IP architecture. An end-to-end IP 1352 architecture takes advantage of nearly three decades of IP technology 1353 development, facilitating interoperability and device management 1354 across disparate networks and devices, as it has been already 1355 demonstrated in many mission-critical and highly secure networks. 1357 IPv6 is seen as a future telecommunications technology for the Smart 1358 Grid; the IEC (International Electrotechnical Commission) and 1359 different National Committees have mandated a specific adhoc group 1360 (AHG8) to define the migration strategy to IPv6 for all the IEC TC57 1361 power automation standards. The AHG8 has recently finalised the work 1362 on the migration strategy and the following Technical Report has been 1363 issued: IEC TR 62357-200:2015: Guidelines for migration from Internet 1364 Protocol version 4 (IPv4) to Internet Protocol version 6 (IPv6). 1366 We expect cloud-based SCADA systems to control and monitor the 1367 critical and non-critical subsystems of generation systems, for 1368 example wind farms. 1370 3.3.1. Migration to Packet-Switched Network 1372 Throughout the world, utilities are increasingly planning for a 1373 future based on smart grid applications requiring advanced 1374 telecommunications systems. Many of these applications utilize 1375 packet connectivity for communicating information and control signals 1376 across the utility's Wide Area Network (WAN), made possible by 1377 technologies such as multiprotocol label switching (MPLS). The data 1378 that traverses the utility WAN includes: 1380 o Grid monitoring, control, and protection data 1382 o Non-control grid data (e.g. asset data for condition-based 1383 monitoring) 1385 o Physical safety and security data (e.g. voice and video) 1387 o Remote worker access to corporate applications (voice, maps, 1388 schematics, etc.) 1390 o Field area network backhaul for smart metering, and distribution 1391 grid management 1393 o Enterprise traffic (email, collaboration tools, business 1394 applications) 1396 WANs support this wide variety of traffic to and from substations, 1397 the transmission and distribution grid, generation sites, between 1398 control centers, and between work locations and data centers. To 1399 maintain this rapidly expanding set of applications, many utilities 1400 are taking steps to evolve present time-division multiplexing (TDM) 1401 based and frame relay infrastructures to packet systems. Packet- 1402 based networks are designed to provide greater functionalities and 1403 higher levels of service for applications, while continuing to 1404 deliver reliability and deterministic (real-time) traffic support. 1406 3.3.2. Telecommunications Trends 1408 These general telecommunications topics are in addition to the use 1409 cases that have been addressed so far. These include both current 1410 and future telecommunications related topics that should be factored 1411 into the network architecture and design. 1413 3.3.2.1. General Telecommunications Requirements 1415 o IP Connectivity everywhere 1417 o Monitoring services everywhere and from different remote centers 1418 o Move services to a virtual data center 1420 o Unify access to applications / information from the corporate 1421 network 1423 o Unify services 1425 o Unified Communications Solutions 1427 o Mix of fiber and microwave technologies - obsolescence of SONET/ 1428 SDH or TDM 1430 o Standardize grid telecommunications protocol to opened standard to 1431 ensure interoperability 1433 o Reliable Telecommunications for Transmission and Distribution 1434 Substations 1436 o IEEE 1588 time synchronization Client / Server Capabilities 1438 o Integration of Multicast Design 1440 o QoS Requirements Mapping 1442 o Enable Future Network Expansion 1444 o Substation Network Resilience 1446 o Fast Convergence Design 1448 o Scalable Headend Design 1450 o Define Service Level Agreements (SLA) and Enable SLA Monitoring 1452 o Integration of 3G/4G Technologies and future technologies 1454 o Ethernet Connectivity for Station Bus Architecture 1456 o Ethernet Connectivity for Process Bus Architecture 1458 o Protection, teleprotection and PMU (Phaser Measurement Unit) on IP 1460 3.3.2.2. Specific Network topologies of Smart Grid Applications 1462 Utilities often have very large private telecommunications networks. 1463 It covers an entire territory / country. The main purpose of the 1464 network, until now, has been to support transmission network 1465 monitoring, control, and automation, remote control of generation 1466 sites, and providing FCAPS (Fault, Configuration, Accounting, 1467 Performance, Security) services from centralized network operation 1468 centers. 1470 Going forward, one network will support operation and maintenance of 1471 electrical networks (generation, transmission, and distribution), 1472 voice and data services for ten of thousands of employees and for 1473 exchange with neighboring interconnections, and administrative 1474 services. To meet those requirements, utility may deploy several 1475 physical networks leveraging different technologies across the 1476 country: an optical network and a microwave network for instance. 1477 Each protection and automatism system between two points has two 1478 telecommunications circuits, one on each network. Path diversity 1479 between two substations is key. Regardless of the event type 1480 (hurricane, ice storm, etc.), one path shall stay available so the 1481 system can still operate. 1483 In the optical network, signals are transmitted over more than tens 1484 of thousands of circuits using fiber optic links, microwave and 1485 telephone cables. This network is the nervous system of the 1486 utility's power transmission operations. The optical network 1487 represents ten of thousands of km of cable deployed along the power 1488 lines, with individual runs as long as 280 km. 1490 3.3.2.3. Precision Time Protocol 1492 Some utilities do not use GPS clocks in generation substations. One 1493 of the main reasons is that some of the generation plants are 30 to 1494 50 meters deep under ground and the GPS signal can be weak and 1495 unreliable. Instead, atomic clocks are used. Clocks are 1496 synchronized amongst each other. Rubidium clocks provide clock and 1497 1ms timestamps for IRIG-B. 1499 Some companies plan to transition to the Precision Time Protocol 1500 (PTP, [IEEE1588]), distributing the synchronization signal over the 1501 IP/MPLS network. PTP provides a mechanism for synchronizing the 1502 clocks of participating nodes to a high degree of accuracy and 1503 precision. 1505 PTP operates based on the following assumptions: 1507 It is assumed that the network eliminates cyclic forwarding of PTP 1508 messages within each communication path (e.g. by using a spanning 1509 tree protocol). 1511 PTP is tolerant of an occasional missed message, duplicated 1512 message, or message that arrived out of order. However, PTP 1513 assumes that such impairments are relatively rare. 1515 PTP was designed assuming a multicast communication model, however 1516 PTP also supports a unicast communication model as long as the 1517 behavior of the protocol is preserved. 1519 Like all message-based time transfer protocols, PTP time accuracy 1520 is degraded by delay asymmetry in the paths taken by event 1521 messages. Asymmetry is not detectable by PTP, however, if such 1522 delays are known a priori, PTP can correct for asymmetry. 1524 IEC 61850 defines the use of IEC/IEEE 61850-9-3:2016. The title is: 1525 Precision time protocol profile for power utility automation. It is 1526 based on Annex B/IEC 62439 which offers the support of redundant 1527 attachment of clocks to Parallel Redundancy Protocol (PRP) and High- 1528 availability Seamless Redundancy (HSR) networks. 1530 3.3.3. Security Trends in Utility Networks 1532 Although advanced telecommunications networks can assist in 1533 transforming the energy industry by playing a critical role in 1534 maintaining high levels of reliability, performance, and 1535 manageability, they also introduce the need for an integrated 1536 security infrastructure. Many of the technologies being deployed to 1537 support smart grid projects such as smart meters and sensors can 1538 increase the vulnerability of the grid to attack. Top security 1539 concerns for utilities migrating to an intelligent smart grid 1540 telecommunications platform center on the following trends: 1542 o Integration of distributed energy resources 1544 o Proliferation of digital devices to enable management, automation, 1545 protection, and control 1547 o Regulatory mandates to comply with standards for critical 1548 infrastructure protection 1550 o Migration to new systems for outage management, distribution 1551 automation, condition-based maintenance, load forecasting, and 1552 smart metering 1554 o Demand for new levels of customer service and energy management 1556 This development of a diverse set of networks to support the 1557 integration of microgrids, open-access energy competition, and the 1558 use of network-controlled devices is driving the need for a converged 1559 security infrastructure for all participants in the smart grid, 1560 including utilities, energy service providers, large commercial and 1561 industrial, as well as residential customers. Securing the assets of 1562 electric power delivery systems (from the control center to the 1563 substation, to the feeders and down to customer meters) requires an 1564 end-to-end security infrastructure that protects the myriad of 1565 telecommunications assets used to operate, monitor, and control power 1566 flow and measurement. 1568 "Cyber security" refers to all the security issues in automation and 1569 telecommunications that affect any functions related to the operation 1570 of the electric power systems. Specifically, it involves the 1571 concepts of: 1573 o Integrity : data cannot be altered undetectably 1575 o Authenticity : the telecommunications parties involved must be 1576 validated as genuine 1578 o Authorization : only requests and commands from the authorized 1579 users can be accepted by the system 1581 o Confidentiality : data must not be accessible to any 1582 unauthenticated users 1584 When designing and deploying new smart grid devices and 1585 telecommunications systems, it is imperative to understand the 1586 various impacts of these new components under a variety of attack 1587 situations on the power grid. Consequences of a cyber attack on the 1588 grid telecommunications network can be catastrophic. This is why 1589 security for smart grid is not just an ad hoc feature or product, 1590 it's a complete framework integrating both physical and Cyber 1591 security requirements and covering the entire smart grid networks 1592 from generation to distribution. Security has therefore become one 1593 of the main foundations of the utility telecom network architecture 1594 and must be considered at every layer with a defense-in-depth 1595 approach. Migrating to IP based protocols is key to address these 1596 challenges for two reasons: 1598 o IP enables a rich set of features and capabilities to enhance the 1599 security posture 1601 o IP is based on open standards, which allows interoperability 1602 between different vendors and products, driving down the costs 1603 associated with implementing security solutions in OT networks. 1605 Securing OT (Operation technology) telecommunications over packet- 1606 switched IP networks follow the same principles that are foundational 1607 for securing the IT infrastructure, i.e., consideration must be given 1608 to enforcing electronic access control for both person-to-machine and 1609 machine-to-machine communications, and providing the appropriate 1610 levels of data privacy, device and platform integrity, and threat 1611 detection and mitigation. 1613 3.4. Electrical Utilities Asks 1615 o Mixed L2 and L3 topologies 1617 o Deterministic behavior 1619 o Bounded latency and jitter 1621 o Tight feedback intervals 1623 o High availability, low recovery time 1625 o Redundancy, low packet loss 1627 o Precise timing 1629 o Centralized computing of deterministic paths 1631 o Distributed configuration may also be useful 1633 4. Building Automation Systems 1635 4.1. Use Case Description 1637 A Building Automation System (BAS) manages equipment and sensors in a 1638 building for improving residents' comfort, reducing energy 1639 consumption, and responding to failures and emergencies. For 1640 example, the BAS measures the temperature of a room using sensors and 1641 then controls the HVAC (heating, ventilating, and air conditioning) 1642 to maintain a set temperature and minimize energy consumption. 1644 A BAS primarily performs the following functions: 1646 o Periodically measures states of devices, for example humidity and 1647 illuminance of rooms, open/close state of doors, FAN speed, etc. 1649 o Stores the measured data. 1651 o Provides the measured data to BAS systems and operators. 1653 o Generates alarms for abnormal state of devices. 1655 o Controls devices (e.g. turn off room lights at 10:00 PM). 1657 4.2. Building Automation Systems Today 1659 4.2.1. BAS Architecture 1661 A typical BAS architecture of today is shown in Figure 4. 1663 +----------------------------+ 1664 | | 1665 | BMS HMI | 1666 | | | | 1667 | +----------------------+ | 1668 | | Management Network | | 1669 | +----------------------+ | 1670 | | | | 1671 | LC LC | 1672 | | | | 1673 | +----------------------+ | 1674 | | Field Network | | 1675 | +----------------------+ | 1676 | | | | | | 1677 | Dev Dev Dev Dev | 1678 | | 1679 +----------------------------+ 1681 BMS := Building Management Server 1682 HMI := Human Machine Interface 1683 LC := Local Controller 1685 Figure 4: BAS architecture 1687 There are typically two layers of network in a BAS. The upper one is 1688 called the Management Network and the lower one is called the Field 1689 Network. In management networks an IP-based communication protocol 1690 is used, while in field networks non-IP based communication protocols 1691 ("field protocols") are mainly used. Field networks have specific 1692 timing requirements, whereas management networks can be best-effort. 1694 A Human Machine Interface (HMI) is typically a desktop PC used by 1695 operators to monitor and display device states, send device control 1696 commands to Local Controllers (LCs), and configure building schedules 1697 (for example "turn off all room lights in the building at 10:00 PM"). 1699 A Building Management Server (BMS) performs the following operations. 1701 o Collect and store device states from LCs at regular intervals. 1703 o Send control values to LCs according to a building schedule. 1705 o Send an alarm signal to operators if it detects abnormal devices 1706 states. 1708 The BMS and HMI communicate with LCs via IP-based "management 1709 protocols" (see standards [bacnetip], [knx]). 1711 A LC is typically a Programmable Logic Controller (PLC) which is 1712 connected to several tens or hundreds of devices using "field 1713 protocols". An LC performs the following kinds of operations: 1715 o Measure device states and provide the information to BMS or HMI. 1717 o Send control values to devices, unilaterally or as part of a 1718 feedback control loop. 1720 There are many field protocols used today; some are standards-based 1721 and others are proprietary (see standards [lontalk], [modbus], 1722 [profibus] and [flnet]). The result is that BASs have multiple MAC/ 1723 PHY modules and interfaces. This makes BASs more expensive, slower 1724 to develop, and can result in "vendor lock-in" with multiple types of 1725 management applications. 1727 4.2.2. BAS Deployment Model 1729 An example BAS for medium or large buildings is shown in Figure 5. 1730 The physical layout spans multiple floors, and there is a monitoring 1731 room where the BAS management entities are located. Each floor will 1732 have one or more LCs depending upon the number of devices connected 1733 to the field network. 1735 +--------------------------------------------------+ 1736 | Floor 3 | 1737 | +----LC~~~~+~~~~~+~~~~~+ | 1738 | | | | | | 1739 | | Dev Dev Dev | 1740 | | | 1741 |--- | ------------------------------------------| 1742 | | Floor 2 | 1743 | +----LC~~~~+~~~~~+~~~~~+ Field Network | 1744 | | | | | | 1745 | | Dev Dev Dev | 1746 | | | 1747 |--- | ------------------------------------------| 1748 | | Floor 1 | 1749 | +----LC~~~~+~~~~~+~~~~~+ +-----------------| 1750 | | | | | | Monitoring Room | 1751 | | Dev Dev Dev | | 1752 | | | BMS HMI | 1753 | | Management Network | | | | 1754 | +--------------------------------+-----+ | 1755 | | | 1756 +--------------------------------------------------+ 1758 Figure 5: BAS Deployment model for Medium/Large Buildings 1760 Each LC is connected to the monitoring room via the Management 1761 network, and the management functions are performed within the 1762 building. In most cases, fast Ethernet (e.g. 100BASE-T) is used for 1763 the management network. Since the management network is non- 1764 realtime, use of Ethernet without quality of service is sufficient 1765 for today's deployment. 1767 In the field network a variety of physical interfaces such as RS232C 1768 and RS485 are used, which have specific timing requirements. Thus if 1769 a field network is to be replaced with an Ethernet or wireless 1770 network, such networks must support time-critical deterministic 1771 flows. 1773 In Figure 6, another deployment model is presented in which the 1774 management system is hosted remotely. This is becoming popular for 1775 small office and residential buildings in which a standalone 1776 monitoring system is not cost-effective. 1778 +---------------+ 1779 | Remote Center | 1780 | | 1781 | BMS HMI | 1782 +------------------------------------+ | | | | 1783 | Floor 2 | | +---+---+ | 1784 | +----LC~~~~+~~~~~+ Field Network| | | | 1785 | | | | | | Router | 1786 | | Dev Dev | +-------|-------+ 1787 | | | | 1788 |--- | ------------------------------| | 1789 | | Floor 1 | | 1790 | +----LC~~~~+~~~~~+ | | 1791 | | | | | | 1792 | | Dev Dev | | 1793 | | | | 1794 | | Management Network | WAN | 1795 | +------------------------Router-------------+ 1796 | | 1797 +------------------------------------+ 1799 Figure 6: Deployment model for Small Buildings 1801 Some interoperability is possible today in the Management Network, 1802 but not in today's field networks due to their non-IP-based design. 1804 4.2.3. Use Cases for Field Networks 1806 Below are use cases for Environmental Monitoring, Fire Detection, and 1807 Feedback Control, and their implications for field network 1808 performance. 1810 4.2.3.1. Environmental Monitoring 1812 The BMS polls each LC at a maximum measurement interval of 100ms (for 1813 example to draw a historical chart of 1 second granularity with a 10x 1814 sampling interval) and then performs the operations as specified by 1815 the operator. Each LC needs to measure each of its several hundred 1816 sensors once per measurement interval. Latency is not critical in 1817 this scenario as long as all sensor values are completed in the 1818 measurement interval. Availability is expected to be 99.999 %. 1820 4.2.3.2. Fire Detection 1822 On detection of a fire, the BMS must stop the HVAC, close the fire 1823 shutters, turn on the fire sprinklers, send an alarm, etc. There are 1824 typically ~10s of sensors per LC that BMS needs to manage. In this 1825 scenario the measurement interval is 10-50ms, the communication delay 1826 is 10ms, and the availability must be 99.9999 %. 1828 4.2.3.3. Feedback Control 1830 BAS systems utilize feedback control in various ways; the most time- 1831 critial is control of DC motors, which require a short feedback 1832 interval (1-5ms) with low communication delay (10ms) and jitter 1833 (1ms). The feedback interval depends on the characteristics of the 1834 device and a target quality of control value. There are typically 1835 ~10s of such devices per LC. 1837 Communication delay is expected to be less than 10 ms, jitter less 1838 than 1 sec while the availability must be 99.9999% . 1840 4.2.4. Security Considerations 1842 When BAS field networks were developed it was assumed that the field 1843 networks would always be physically isolated from external networks 1844 and therefore security was not a concern. In today's world many BASs 1845 are managed remotely and are thus connected to shared IP networks and 1846 so security is definitely a concern, yet security features are not 1847 available in the majority of BAS field network deployments . 1849 The management network, being an IP-based network, has the protocols 1850 available to enable network security, but in practice many BAS 1851 systems do not implement even the available security features such as 1852 device authentication or encryption for data in transit. 1854 4.3. BAS Future 1856 In the future we expect more fine-grained environmental monitoring 1857 and lower energy consumption, which will require more sensors and 1858 devices, thus requiring larger and more complex building networks. 1860 We expect building networks to be connected to or converged with 1861 other networks (Enterprise network, Home network, and Internet). 1863 Therefore better facilities for network management, control, 1864 reliability and security are critical in order to improve resident 1865 and operator convenience and comfort. For example the ability to 1866 monitor and control building devices via the internet would enable 1867 (for example) control of room lights or HVAC from a resident's 1868 desktop PC or phone application. 1870 4.4. BAS Asks 1872 The community would like to see an interoperable protocol 1873 specification that can satisfy the timing, security, availability and 1874 QoS constraints described above, such that the resulting converged 1875 network can replace the disparate field networks. Ideally this 1876 connectivity could extend to the open Internet. 1878 This would imply an architecture that can guarantee 1880 o Low communication delays (from <10ms to 100ms in a network of 1881 several hundred devices) 1883 o Low jitter (< 1 ms) 1885 o Tight feedback intervals (1ms - 10ms) 1887 o High network availability (up to 99.9999% ) 1889 o Availability of network data in disaster scenario 1891 o Authentication between management and field devices (both local 1892 and remote) 1894 o Integrity and data origin authentication of communication data 1895 between field and management devices 1897 o Confidentiality of data when communicated to a remote device 1899 5. Wireless for Industrial 1901 5.1. Use Case Description 1903 Wireless networks are useful for industrial applications, for example 1904 when portable, fast-moving or rotating objects are involved, and for 1905 the resource-constrained devices found in the Internet of Things 1906 (IoT). 1908 Such network-connected sensors, actuators, control loops (etc.) 1909 typically require that the underlying network support real-time 1910 quality of service (QoS), as well as specific classes of other 1911 network properties such as reliability, redundancy, and security. 1913 These networks may also contain very large numbers of devices, for 1914 example for factories, "big data" acquisition, and the IoT. Given 1915 the large numbers of devices installed, and the potential 1916 pervasiveness of the IoT, this is a huge and very cost-sensitive 1917 market. For example, a 1% cost reduction in some areas could save 1918 $100B 1920 5.1.1. Network Convergence using 6TiSCH 1922 Some wireless network technologies support real-time QoS, and are 1923 thus useful for these kinds of networks, but others do not. For 1924 example WiFi is pervasive but does not provide guaranteed timing or 1925 delivery of packets, and thus is not useful in this context. 1927 In this use case we focus on one specific wireless network technology 1928 which does provide the required deterministic QoS, which is "IPv6 1929 over the TSCH mode of IEEE 802.15.4e" (6TiSCH, where TSCH stands for 1930 "Time-Slotted Channel Hopping", see [I-D.ietf-6tisch-architecture], 1931 [IEEE802154], [IEEE802154e], and [RFC7554]). 1933 There are other deterministic wireless busses and networks available 1934 today, however they are imcompatible with each other, and 1935 incompatible with IP traffic (for example [ISA100], [WirelessHART]). 1937 Thus the primary goal of this use case is to apply 6TiSH as a 1938 converged IP- and standards-based wireless network for industrial 1939 applications, i.e. to replace multiple proprietary and/or 1940 incompatible wireless networking and wireless network management 1941 standards. 1943 5.1.2. Common Protocol Development for 6TiSCH 1945 Today there are a number of protocols required by 6TiSCH which are 1946 still in development, and a second intent of this use case is to 1947 highlight the ways in which these "missing" protocols share goals in 1948 common with DetNet. Thus it is possible that some of the protocol 1949 technology developed for DetNet will also be applicable to 6TiSCH. 1951 These protocol goals are identified here, along with their 1952 relationship to DetNet. It is likely that ultimately the resulting 1953 protocols will not be identical, but will share design principles 1954 which contribute to the eficiency of enabling both DetNet and 6TiSCH. 1956 One such commonality is that although at a different time scale, in 1957 both TSN [IEEE802.1TSNTG] and TSCH a packet crosses the network from 1958 node to node follows a precise schedule, as a train that leaves 1959 intermediate stations at precise times along its path. This kind of 1960 operation reduces collisions, saves energy, and enables engineering 1961 the network for deterministic properties. 1963 Another commonality is remote monitoring and scheduling management of 1964 a TSCH network by a Path Computation Element (PCE) and Network 1965 Management Entity (NME). The PCE/NME manage timeslots and device 1966 resources in a manner that minimizes the interaction with and the 1967 load placed on resource-constrained devices. For example, a tiny IoT 1968 device may have just enough buffers to store one or a few IPv6 1969 packets, and will have limited bandwidth between peers such that it 1970 can maintain only a small amount of peer information, and will not be 1971 able to store many packets waiting to be forwarded. It is 1972 advantageous then for it to only be required to carry out the 1973 specific behavior assigned to it by the PCE/NME (as opposed to 1974 maintaining its own IP stack, for example). 1976 Note: Current WG discussion indicates that some peer-to-peer 1977 communication must be assumed, i.e. the PCE may communicate only 1978 indirectly with any given device, enabling hierarchical configuration 1979 of the system. 1981 6TiSCH depends on [PCE] and [I-D.finn-detnet-architecture]. 1983 6TiSCH also depends on the fact that DetNet will maintain consistency 1984 with [IEEE802.1TSNTG]. 1986 5.2. Wireless Industrial Today 1988 Today industrial wireless is accomplished using multiple 1989 deterministic wireless networks which are incompatible with each 1990 other and with IP traffic. 1992 6TiSCH is not yet fully specified, so it cannot be used in today's 1993 applications. 1995 5.3. Wireless Industrial Future 1997 5.3.1. Unified Wireless Network and Management 1999 We expect DetNet and 6TiSCH together to enable converged transport of 2000 deterministic and best-effort traffic flows between real-time 2001 industrial devices and wide area networks via IP routing. A high 2002 level view of a basic such network is shown in Figure 7. 2004 ---+-------- ............ ------------ 2005 | External Network | 2006 | +-----+ 2007 +-----+ | NME | 2008 | | LLN Border | | 2009 | | router +-----+ 2010 +-----+ 2011 o o o 2012 o o o o 2013 o o LLN o o o 2014 o o o o 2015 o 2017 Figure 7: Basic 6TiSCH Network 2019 Figure 8 shows a backbone router federating multiple synchronized 2020 6TiSCH subnets into a single subnet connected to the external 2021 network. 2023 ---+-------- ............ ------------ 2024 | External Network | 2025 | +-----+ 2026 | +-----+ | NME | 2027 +-----+ | +-----+ | | 2028 | | Router | | PCE | +-----+ 2029 | | +--| | 2030 +-----+ +-----+ 2031 | | 2032 | Subnet Backbone | 2033 +--------------------+------------------+ 2034 | | | 2035 +-----+ +-----+ +-----+ 2036 | | Backbone | | Backbone | | Backbone 2037 o | | router | | router | | router 2038 +-----+ +-----+ +-----+ 2039 o o o o o 2040 o o o o o o o o o o o 2041 o o o LLN o o o o 2042 o o o o o o o o o o o o 2044 Figure 8: Extended 6TiSCH Network 2046 The backbone router must ensure end-to-end deterministic behavior 2047 between the LLN and the backbone. We would like to see this 2048 accomplished in conformance with the work done in 2049 [I-D.finn-detnet-architecture] with respect to Layer-3 aspects of 2050 deterministic networks that span multiple Layer-2 domains. 2052 The PCE must compute a deterministic path end-to-end across the TSCH 2053 network and IEEE802.1 TSN Ethernet backbone, and DetNet protocols are 2054 expected to enable end-to-end deterministic forwarding. 2056 +-----+ 2057 | IoT | 2058 | G/W | 2059 +-----+ 2060 ^ <---- Elimination 2061 | | 2062 Track branch | | 2063 +-------+ +--------+ Subnet Backbone 2064 | | 2065 +--|--+ +--|--+ 2066 | | | Backbone | | | Backbone 2067 o | | | router | | | router 2068 +--/--+ +--|--+ 2069 o / o o---o----/ o 2070 o o---o--/ o o o o o 2071 o \ / o o LLN o 2072 o v <---- Replication 2073 o 2075 Figure 9: 6TiSCH Network with PRE 2077 5.3.1.1. PCE and 6TiSCH ARQ Retries 2079 Note: The possible use of ARQ techniques in DetNet is currently 2080 considered a possible design alternative. 2082 6TiSCH uses the IEEE802.15.4 Automatic Repeat-reQuest (ARQ) mechanism 2083 to provide higher reliability of packet delivery. ARQ is related to 2084 packet replication and elimination because there are two independent 2085 paths for packets to arrive at the destination, and if an expected 2086 packed does not arrive on one path then it checks for the packet on 2087 the second path. 2089 Although to date this mechanism is only used by wireless networks, 2090 this may be a technique that would be appropriate for DetNet and so 2091 aspects of the enabling protocol could be co-developed. 2093 For example, in Figure 9, a Track is laid out from a field device in 2094 a 6TiSCH network to an IoT gateway that is located on a IEEE802.1 TSN 2095 backbone. 2097 In ARQ the Replication function in the field device sends a copy of 2098 each packet over two different branches, and the PCE schedules each 2099 hop of both branches so that the two copies arrive in due time at the 2100 gateway. In case of a loss on one branch, hopefully the other copy 2101 of the packet still arrives within the allocated time. If two copies 2102 make it to the IoT gateway, the Elimination function in the gateway 2103 ignores the extra packet and presents only one copy to upper layers. 2105 At each 6TiSCH hop along the Track, the PCE may schedule more than 2106 one timeSlot for a packet, so as to support Layer-2 retries (ARQ). 2108 In current deployments, a TSCH Track does not necessarily support PRE 2109 but is systematically multi-path. This means that a Track is 2110 scheduled so as to ensure that each hop has at least two forwarding 2111 solutions, and the forwarding decision is to try the preferred one 2112 and use the other in case of Layer-2 transmission failure as detected 2113 by ARQ. 2115 5.3.2. Schedule Management by a PCE 2117 A common feature of 6TiSCH and DetNet is the action of a PCE to 2118 configure paths through the network. Specifically, what is needed is 2119 a protocol and data model that the PCE will use to get/set the 2120 relevant configuration from/to the devices, as well as perform 2121 operations on the devices. We expect that this protocol will be 2122 developed by DetNet with consideration for its reuse by 6TiSCH. The 2123 remainder of this section provides a bit more context from the 6TiSCH 2124 side. 2126 5.3.2.1. PCE Commands and 6TiSCH CoAP Requests 2128 The 6TiSCH device does not expect to place the request for bandwidth 2129 between itself and another device in the network. Rather, an 2130 operation control system invoked through a human interface specifies 2131 the required traffic specification and the end nodes (in terms of 2132 latency and reliability). Based on this information, the PCE must 2133 compute a path between the end nodes and provision the network with 2134 per-flow state that describes the per-hop operation for a given 2135 packet, the corresponding timeslots, and the flow identification that 2136 enables recognizing that a certain packet belongs to a certain path, 2137 etc. 2139 For a static configuration that serves a certain purpose for a long 2140 period of time, it is expected that a node will be provisioned in one 2141 shot with a full schedule, which incorporates the aggregation of its 2142 behavior for multiple paths. 6TiSCH expects that the programing of 2143 the schedule will be done over COAP as discussed in 2144 [I-D.ietf-6tisch-coap]. 2146 6TiSCH expects that the PCE commands will be mapped back and forth 2147 into CoAP by a gateway function at the edge of the 6TiSCH network. 2148 For instance, it is possible that a mapping entity on the backbone 2149 transforms a non-CoAP protocol such as PCEP into the RESTful 2150 interfaces that the 6TiSCH devices support. This architecture will 2151 be refined to comply with DetNet [I-D.finn-detnet-architecture] when 2152 the work is formalized. Related information about 6TiSCH can be 2153 found at [I-D.ietf-6tisch-6top-interface] and RPL [RFC6550]. 2155 A protocol may be used to update the state in the devices during 2156 runtime, for example if it appears that a path through the network 2157 has ceased to perform as expected, but in 6TiSCH that flow was not 2158 designed and no protocol was selected. We would like to see DetNet 2159 define the appropriate end-to-end protocols to be used in that case. 2160 The implication is that these state updates take place once the 2161 system is configured and running, i.e. they are not limited to the 2162 initial communication of the configuration of the system. 2164 A "slotFrame" is the base object that a PCE would manipulate to 2165 program a schedule into an LLN node ([I-D.ietf-6tisch-architecture]). 2167 We would like to see the PCE read energy data from devices, and 2168 compute paths that will implement policies on how energy in devices 2169 is consumed, for instance to ensure that the spent energy does not 2170 exceeded the available energy over a period of time. Note: this 2171 statement implies that an extensible protocol for communicating 2172 device info to the PCE and enabling the PCE to act on it will be part 2173 of the DetNet architecture, however for subnets with specific 2174 protocols (e.g. CoAP) a gateway may be required. 2176 6TiSCH devices can discover their neighbors over the radio using a 2177 mechanism such as beacons, but even though the neighbor information 2178 is available in the 6TiSCH interface data model, 6TiSCH does not 2179 describe a protocol to proactively push the neighborhood information 2180 to a PCE. We would like to see DetNet define such a protocol; one 2181 possible design alternative is that it could operate over CoAP, 2182 alternatively it could be converted to/from CoAP by a gateway. We 2183 would like to see such a protocol carry multiple metrics, for example 2184 similar to those used for RPL operations [RFC6551] 2186 5.3.2.2. 6TiSCH IP Interface 2188 "6top" ([I-D.wang-6tisch-6top-sublayer]) is a logical link control 2189 sitting between the IP layer and the TSCH MAC layer which provides 2190 the link abstraction that is required for IP operations. The 6top 2191 data model and management interfaces are further discussed in 2192 [I-D.ietf-6tisch-6top-interface] and [I-D.ietf-6tisch-coap]. 2194 An IP packet that is sent along a 6TiSCH path uses the Differentiated 2195 Services Per-Hop-Behavior Group called Deterministic Forwarding, as 2196 described in [I-D.svshah-tsvwg-deterministic-forwarding]. 2198 5.3.3. 6TiSCH Security Considerations 2200 On top of the classical requirements for protection of control 2201 signaling, it must be noted that 6TiSCH networks operate on limited 2202 resources that can be depleted rapidly in a DoS attack on the system, 2203 for instance by placing a rogue device in the network, or by 2204 obtaining management control and setting up unexpected additional 2205 paths. 2207 5.4. Wireless Industrial Asks 2209 6TiSCH depends on DetNet to define: 2211 o Configuration (state) and operations for deterministic paths 2213 o End-to-end protocols for deterministic forwarding (tagging, IP) 2215 o Protocol for packet replication and elimination 2217 6. Cellular Radio 2219 6.1. Use Case Description 2221 This use case describes the application of deterministic networking 2222 in the context of cellular telecom transport networks. Important 2223 elements include time synchronization, clock distribution, and ways 2224 of establishing time-sensitive streams for both Layer-2 and Layer-3 2225 user plane traffic. 2227 6.1.1. Network Architecture 2229 Figure 10 illustrates a typical 3GPP-defined cellular network 2230 architecture, which includes "Fronthaul", "Midhaul" and "Backhaul" 2231 network segments. The "Fronthaul" is the network connecting base 2232 stations (baseband processing units) to the remote radio heads 2233 (antennas). The "Midhaul" is the network inter-connecting base 2234 stations (or small cell sites). The "Backhaul" is the network or 2235 links connecting the radio base station sites to the network 2236 controller/gateway sites (i.e. the core of the 3GPP cellular 2237 network). 2239 In Figure 10 "eNB" ("E-UTRAN Node B") is the hardware that is 2240 connected to the mobile phone network which communicates directly 2241 with mobile handsets ([TS36300]). 2243 Y (remote radio heads (antennas)) 2244 \ 2245 Y__ \.--. .--. +------+ 2246 \_( `. +---+ _(Back`. | 3GPP | 2247 Y------( Front )----|eNB|----( Haul )----| core | 2248 ( ` .Haul ) +---+ ( ` . ) ) | netw | 2249 /`--(___.-' \ `--(___.-' +------+ 2250 Y_/ / \.--. \ 2251 Y_/ _( Mid`. \ 2252 ( Haul ) \ 2253 ( ` . ) ) \ 2254 `--(___.-'\_____+---+ (small cell sites) 2255 \ |SCe|__Y 2256 +---+ +---+ 2257 Y__|eNB|__Y 2258 +---+ 2259 Y_/ \_Y ("local" radios) 2261 Figure 10: Generic 3GPP-based Cellular Network Architecture 2263 6.1.2. Delay Constraints 2265 The available processing time for Fronthaul networking overhead is 2266 limited to the available time after the baseband processing of the 2267 radio frame has completed. For example in Long Term Evolution (LTE) 2268 radio, processing of a radio frame is allocated 3ms but typically the 2269 processing uses most of it, allowing only a small fraction to be used 2270 by the Fronthaul network (e.g. up to 250us one-way delay, though the 2271 existing spec ([NGMN-fronth]) supports delay only up to 100us). This 2272 ultimately determines the distance the remote radio heads can be 2273 located from the base stations (e.g., 100us equals roughly 20 km of 2274 optical fiber-based transport). Allocation options of the available 2275 time budget between processing and transport are under heavy 2276 discussions in the mobile industry. 2278 For packet-based transport the allocated transport time (e.g. CPRI 2279 would allow for 100us delay [CPRI]) is consumed by all nodes and 2280 buffering between the remote radio head and the baseband processing 2281 unit, plus the distance-incurred delay. 2283 The baseband processing time and the available "delay budget" for the 2284 fronthaul is likely to change in the forthcoming "5G" due to reduced 2285 radio round trip times and other architectural and service 2286 requirements [NGMN]. 2288 The transport time budget, as noted above, places limitations on the 2289 distance that remote radio heads can be located from base stations 2290 (i.e. the link length). In the above analysis, the entire transport 2291 time budget is assumed to be available for link propagation delay. 2292 However the transport time budget can be broken down into three 2293 components: scheduling /queueing delay, transmission delay, and link 2294 propagation delay. Using today's Fronthaul networking technology, 2295 the queuing, scheduling and transmission components might become the 2296 dominant factors in the total transport time rather than the link 2297 propagation delay. This is especially true in cases where the 2298 Fronthaul link is relatively short and it is shared among multiple 2299 Fronthaul flows, for example in indoor and small cell networks, 2300 massive MIMO antenna networks, and split Fronthaul architectures. 2302 DetNet technology can improve this application by controlling and 2303 reducing the time required for the queuing, scheduling and 2304 transmission operations by properly assigning the network resources, 2305 thus leaving more of the transport time budget available for link 2306 propagation, and thus enabling longer link lengths. However, link 2307 length is usually a given parameter and is not a controllable network 2308 parameter, since RRH and BBU sights are usually located in 2309 predetermined locations. However, the number of antennas in an RRH 2310 sight might increase for example by adding more antennas, increasing 2311 the MIMO capability of the network or support of massive MIMO. This 2312 means increasing the number of the fronthaul flows sharing the same 2313 fronthaul link. DetNet can now control the bandwidth assignment of 2314 the fronthaul link and the scheduling pf fronthaul packets over this 2315 link and provide adequate buffer provisioning for each flow to reduce 2316 the packet loss rate. 2318 Another way in which DetNet technology can aid Fronthaul networks is 2319 by providing effective isolation from best-effort (and other classes 2320 of) traffic, which can arise as a result of network slicing in 5G 2321 networks where Fronthaul traffic generated in different network 2322 slices might have differing performance requirements. DetNet 2323 technology can also dynamically control the bandwidth assignment, 2324 scheduling and packet forwarding decisions and the buffer 2325 provisioning of the Fronthaul flows to guarantee the end-to-end delay 2326 of the Fronthaul packets and minimize the packet loss rate. 2328 [METIS] documents the fundamental challenges as well as overall 2329 technical goals of the future 5G mobile and wireless system as the 2330 starting point. These future systems should support much higher data 2331 volumes and rates and significantly lower end-to-end latency for 100x 2332 more connected devices (at similar cost and energy consumption levels 2333 as today's system). 2335 For Midhaul connections, delay constraints are driven by Inter-Site 2336 radio functions like Coordinated Multipoint Processing (CoMP, see 2337 [CoMP]). CoMP reception and transmission is a framework in which 2338 multiple geographically distributed antenna nodes cooperate to 2339 improve the performance of the users served in the common cooperation 2340 area. The design principal of CoMP is to extend the current single- 2341 cell to multi-UE (User Equipment) transmission to a multi-cell-to- 2342 multi-UEs transmission by base station cooperation. 2344 CoMP has delay-sensitive performance parameters, which are "midhaul 2345 latency" and "CSI (Channel State Information) reporting and 2346 accuracy". The essential feature of CoMP is signaling between eNBs, 2347 so Midhaul latency is the dominating limitation of CoMP performance. 2348 Generally, CoMP can benefit from coordinated scheduling (either 2349 distributed or centralized) of different cells if the signaling delay 2350 between eNBs is within 1-10ms. This delay requirement is both rigid 2351 and absolute because any uncertainty in delay will degrade the 2352 performance significantly. 2354 Inter-site CoMP is one of the key requirements for 5G and is also a 2355 near-term goal for the current 4.5G network architecture. 2357 6.1.3. Time Synchronization Constraints 2359 Fronthaul time synchronization requirements are given by [TS25104], 2360 [TS36104], [TS36211], and [TS36133]. These can be summarized for the 2361 current 3GPP LTE-based networks as: 2363 Delay Accuracy: 2364 +-8ns (i.e. +-1/32 Tc, where Tc is the UMTS Chip time of 1/3.84 2365 MHz) resulting in a round trip accuracy of +-16ns. The value is 2366 this low to meet the 3GPP Timing Alignment Error (TAE) measurement 2367 requirements. Note: performance guarantees of low nanosecond 2368 values such as these are considered to be below the DetNet layer - 2369 it is assumed that the underlying implementation, e.g. the 2370 hardware, will provide sufficient support (e.g. buffering) to 2371 enable this level of accuracy. These values are maintained in the 2372 use case to give an indication of the overall application. 2374 Timing Alignment Error: 2375 Timing Alignment Error (TAE) is problematic to Fronthaul networks 2376 and must be minimized. If the transport network cannot guarantee 2377 low enough TAE then additional buffering has to be introduced at 2378 the edges of the network to buffer out the jitter. Buffering is 2379 not desirable as it reduces the total available delay budget. 2380 Packet Delay Variation (PDV) requirements can be derived from TAE 2381 for packet based Fronthaul networks. 2383 * For multiple input multiple output (MIMO) or TX diversity 2384 transmissions, at each carrier frequency, TAE shall not exceed 2385 65 ns (i.e. 1/4 Tc). 2387 * For intra-band contiguous carrier aggregation, with or without 2388 MIMO or TX diversity, TAE shall not exceed 130 ns (i.e. 1/2 2389 Tc). 2391 * For intra-band non-contiguous carrier aggregation, with or 2392 without MIMO or TX diversity, TAE shall not exceed 260 ns (i.e. 2393 one Tc). 2395 * For inter-band carrier aggregation, with or without MIMO or TX 2396 diversity, TAE shall not exceed 260 ns. 2398 Transport link contribution to radio frequency error: 2399 +-2 PPB. This value is considered to be "available" for the 2400 Fronthaul link out of the total 50 PPB budget reserved for the 2401 radio interface. Note: the reason that the transport link 2402 contributes to radio frequency error is as follows. The current 2403 way of doing Fronthaul is from the radio unit to remote radio head 2404 directly. The remote radio head is essentially a passive device 2405 (without buffering etc.) The transport drives the antenna 2406 directly by feeding it with samples and everything the transport 2407 adds will be introduced to radio as-is. So if the transport 2408 causes additional frequency error that shows immediately on the 2409 radio as well. Note: performance guarantees of low nanosecond 2410 values such as these are considered to be below the DetNet layer - 2411 it is assumed that the underlying implementation, e.g. the 2412 hardware, will provide sufficient support to enable this level of 2413 performance. These values are maintained in the use case to give 2414 an indication of the overall application. 2416 The above listed time synchronization requirements are difficult to 2417 meet with point-to-point connected networks, and more difficult when 2418 the network includes multiple hops. It is expected that networks 2419 must include buffering at the ends of the connections as imposed by 2420 the jitter requirements, since trying to meet the jitter requirements 2421 in every intermediate node is likely to be too costly. However, 2422 every measure to reduce jitter and delay on the path makes it easier 2423 to meet the end-to-end requirements. 2425 In order to meet the timing requirements both senders and receivers 2426 must remain time synchronized, demanding very accurate clock 2427 distribution, for example support for IEEE 1588 transparent clocks or 2428 boundary clocks in every intermediate node. 2430 In cellular networks from the LTE radio era onward, phase 2431 synchronization is needed in addition to frequency synchronization 2432 ([TS36300], [TS23401]). Time constraints are also important due to 2433 their impact on packet loss. If a packet is delivered too late, then 2434 the packet may be dropped by the host. 2436 6.1.4. Transport Loss Constraints 2438 Fronthaul and Midhaul networks assume almost error-free transport. 2439 Errors can result in a reset of the radio interfaces, which can cause 2440 reduced throughput or broken radio connectivity for mobile customers. 2442 For packetized Fronthaul and Midhaul connections packet loss may be 2443 caused by BER, congestion, or network failure scenarios. Different 2444 fronthaul functional splits are being considered by 3GPP, requiring 2445 strict frame loss ratio (FLR) guarantees. As one example (referring 2446 to the legacy CPRI split which is option 8 in 3GPP) lower layers 2447 splits may imply an FLR of less than 10E-7 for data traffic and less 2448 than 10E-6 for control and management traffic. Current tools for 2449 eliminating packet loss for Fronthaul and Midhaul networks have 2450 serious challenges, for example retransmitting lost packets and/or 2451 using forward error correction (FEC) to circumvent bit errors is 2452 practically impossible due to the additional delay incurred. Using 2453 redundant streams for better guarantees for delivery is also 2454 practically impossible in many cases due to high bandwidth 2455 requirements of Fronthaul and Midhaul networks. Protection switching 2456 is also a candidate but current technologies for the path switch are 2457 too slow to avoid reset of mobile interfaces. 2459 Fronthaul links are assumed to be symmetric, and all Fronthaul 2460 streams (i.e. those carrying radio data) have equal priority and 2461 cannot delay or pre-empt each other. This implies that the network 2462 must guarantee that each time-sensitive flow meets their schedule. 2464 6.1.5. Security Considerations 2466 Establishing time-sensitive streams in the network entails reserving 2467 networking resources for long periods of time. It is important that 2468 these reservation requests be authenticated to prevent malicious 2469 reservation attempts from hostile nodes (or accidental 2470 misconfiguration). This is particularly important in the case where 2471 the reservation requests span administrative domains. Furthermore, 2472 the reservation information itself should be digitally signed to 2473 reduce the risk of a legitimate node pushing a stale or hostile 2474 configuration into another networking node. 2476 Note: This is considered important for the security policy of the 2477 network, but does not affect the core DetNet architecture and design. 2479 6.2. Cellular Radio Networks Today 2481 6.2.1. Fronthaul 2483 Today's Fronthaul networks typically consist of: 2485 o Dedicated point-to-point fiber connection is common 2487 o Proprietary protocols and framings 2489 o Custom equipment and no real networking 2491 Current solutions for Fronthaul are direct optical cables or 2492 Wavelength-Division Multiplexing (WDM) connections. 2494 6.2.2. Midhaul and Backhaul 2496 Today's Midhaul and Backhaul networks typically consist of: 2498 o Mostly normal IP networks, MPLS-TP, etc. 2500 o Clock distribution and sync using 1588 and SyncE 2502 Telecommunication networks in the Mid- and Backhaul are already 2503 heading towards transport networks where precise time synchronization 2504 support is one of the basic building blocks. While the transport 2505 networks themselves have practically transitioned to all-IP packet- 2506 based networks to meet the bandwidth and cost requirements, highly 2507 accurate clock distribution has become a challenge. 2509 In the past, Mid- and Backhaul connections were typically based on 2510 Time Division Multiplexing (TDM-based) and provided frequency 2511 synchronization capabilities as a part of the transport media. 2512 Alternatively other technologies such as Global Positioning System 2513 (GPS) or Synchronous Ethernet (SyncE) are used [SyncE]. 2515 Both Ethernet and IP/MPLS [RFC3031] (and PseudoWires (PWE) [RFC3985] 2516 for legacy transport support) have become popular tools to build and 2517 manage new all-IP Radio Access Networks (RANs) 2518 [I-D.kh-spring-ip-ran-use-case]. Although various timing and 2519 synchronization optimizations have already been proposed and 2520 implemented including 1588 PTP enhancements 2521 [I-D.ietf-tictoc-1588overmpls] and [I-D.ietf-mpls-residence-time], 2522 these solution are not necessarily sufficient for the forthcoming RAN 2523 architectures nor do they guarantee the more stringent time- 2524 synchronization requirements such as [CPRI]. 2526 There are also existing solutions for TDM over IP such as [RFC5087] 2527 and [RFC4553], as well as TDM over Ethernet transports such as 2528 [RFC5086]. 2530 6.3. Cellular Radio Networks Future 2532 Future Cellular Radio Networks will be based on a mix of different 2533 xHaul networks (xHaul = front-, mid- and backhaul), and future 2534 transport networks should be able to support all of them 2535 simultaneously. It is already envisioned today that: 2537 o Not all "cellular radio network" traffic will be IP, for example 2538 some will remain at Layer 2 (e.g. Ethernet based). DetNet 2539 solutions must address all traffic types (Layer 2, Layer 3) with 2540 the same tools and allow their transport simultaneously. 2542 o All forms of xHaul networks will need some form of DetNet 2543 solutions. For example with the advent of 5G some Backhaul 2544 traffic will also have DetNet requirements, for example traffic 2545 belonging to time-critical 5G applications. 2547 o Different splits of the functionality run on the base stations and 2548 the on-site units could co-exist on the same Fronthaul and 2549 Backhaul network. 2551 We would like to see the following in future Cellular Radio networks: 2553 o Unified standards-based transport protocols and standard 2554 networking equipment that can make use of underlying deterministic 2555 link-layer services 2557 o Unified and standards-based network management systems and 2558 protocols in all parts of the network (including Fronthaul) 2560 New radio access network deployment models and architectures may 2561 require time- sensitive networking services with strict requirements 2562 on other parts of the network that previously were not considered to 2563 be packetized at all. Time and synchronization support are already 2564 topical for Backhaul and Midhaul packet networks [MEF] and are 2565 becoming a real issue for Fronthaul networks also. Specifically in 2566 Fronthaul networks the timing and synchronization requirements can be 2567 extreme for packet based technologies, for example, on the order of 2568 sub +-20 ns packet delay variation (PDV) and frequency accuracy of 2569 +0.002 PPM [Fronthaul]. 2571 The actual transport protocols and/or solutions to establish required 2572 transport "circuits" (pinned-down paths) for Fronthaul traffic are 2573 still undefined. Those are likely to include (but are not limited 2574 to) solutions directly over Ethernet, over IP, and using MPLS/ 2575 PseudoWire transport. 2577 Even the current time-sensitive networking features may not be 2578 sufficient for Fronthaul traffic. Therefore, having specific 2579 profiles that take the requirements of Fronthaul into account is 2580 desirable [IEEE8021CM]. 2582 Interesting and important work for time-sensitive networking has been 2583 done for Ethernet [TSNTG], which specifies the use of IEEE 1588 time 2584 precision protocol (PTP) [IEEE1588] in the context of IEEE 802.1D and 2585 IEEE 802.1Q. [IEEE8021AS] specifies a Layer 2 time synchronizing 2586 service, and other specifications such as IEEE 1722 [IEEE1722] 2587 specify Ethernet-based Layer-2 transport for time-sensitive streams. 2589 New promising work seeks to enable the transport of time-sensitive 2590 fronthaul streams in Ethernet bridged networks [IEEE8021CM]. 2591 Analogous to IEEE 1722 there is an ongoing standardization effort to 2592 define the Layer-2 transport encapsulation format for transporting 2593 radio over Ethernet (RoE) in the IEEE 1904.3 Task Force [IEEE19043]. 2595 All-IP RANs and xHhaul networks would benefit from time 2596 synchronization and time-sensitive transport services. Although 2597 Ethernet appears to be the unifying technology for the transport, 2598 there is still a disconnect providing Layer 3 services. The protocol 2599 stack typically has a number of layers below the Ethernet Layer 2 2600 that shows up to the Layer 3 IP transport. It is not uncommon that 2601 on top of the lowest layer (optical) transport there is the first 2602 layer of Ethernet followed one or more layers of MPLS, PseudoWires 2603 and/or other tunneling protocols finally carrying the Ethernet layer 2604 visible to the user plane IP traffic. 2606 While there are existing technologies to establish circuits through 2607 the routed and switched networks (especially in MPLS/PWE space), 2608 there is still no way to signal the time synchronization and time- 2609 sensitive stream requirements/reservations for Layer-3 flows in a way 2610 that addresses the entire transport stack, including the Ethernet 2611 layers that need to be configured. 2613 Furthermore, not all "user plane" traffic will be IP. Therefore, the 2614 same solution also must address the use cases where the user plane 2615 traffic is a different layer, for example Ethernet frames. 2617 There is existing work describing the problem statement 2618 [I-D.finn-detnet-problem-statement] and the architecture 2619 [I-D.finn-detnet-architecture] for deterministic networking (DetNet) 2620 that targets solutions for time-sensitive (IP/transport) streams with 2621 deterministic properties over Ethernet-based switched networks. 2623 6.4. Cellular Radio Networks Asks 2625 A standard for data plane transport specification which is: 2627 o Unified among all xHauls (meaning that different flows with 2628 diverse DetNet requirements can coexist in the same network and 2629 traverse the same nodes without interfering with each other) 2631 o Deployed in a highly deterministic network environment 2633 o Capable of supporting multiple functional splits simultaneously, 2634 including existing Backhaul and CPRI Fronthaul and potentially new 2635 modes as defined for example in 3GPP; these goals can be supported 2636 by the existing DetNet Use Case Common Themes, notably "Mix of 2637 Deterministic and Best-Effort Traffic", "Bounded Latency", "Low 2638 Latency", "Symmetrical Path Delays", and "Deterministic Flows". 2640 o Capable of supporting Network Slicing and Multi-tenancy; these 2641 goals can be supported by the same DetNet themes noted above. 2643 o Capable of transporting both in-band and out-band control traffic 2644 (OAM info, ...). 2646 o Deployable over multiple data link technologies (e.g., IEEE 802.3, 2647 mmWave, etc.). 2649 A standard for data flow information models that are: 2651 o Aware of the time sensitivity and constraints of the target 2652 networking environment 2654 o Aware of underlying deterministic networking services (e.g., on 2655 the Ethernet layer) 2657 7. Industrial M2M 2659 7.1. Use Case Description 2661 Industrial Automation in general refers to automation of 2662 manufacturing, quality control and material processing. In this 2663 "machine to machine" (M2M) use case we consider machine units in a 2664 plant floor which periodically exchange data with upstream or 2665 downstream machine modules and/or a supervisory controller within a 2666 local area network. 2668 The actors of M2M communication are Programmable Logic Controllers 2669 (PLCs). Communication between PLCs and between PLCs and the 2670 supervisory PLC (S-PLC) is achieved via critical control/data streams 2671 Figure 11. 2673 S (Sensor) 2674 \ +-----+ 2675 PLC__ \.--. .--. ---| MES | 2676 \_( `. _( `./ +-----+ 2677 A------( Local )-------------( L2 ) 2678 ( Net ) ( Net ) +-------+ 2679 /`--(___.-' `--(___.-' ----| S-PLC | 2680 S_/ / PLC .--. / +-------+ 2681 A_/ \_( `. 2682 (Actuator) ( Local ) 2683 ( Net ) 2684 /`--(___.-'\ 2685 / \ A 2686 S A 2688 Figure 11: Current Generic Industrial M2M Network Architecture 2690 This use case focuses on PLC-related communications; communication to 2691 Manufacturing-Execution-Systems (MESs) are not addressed. 2693 This use case covers only critical control/data streams; non-critical 2694 traffic between industrial automation applications (such as 2695 communication of state, configuration, set-up, and database 2696 communication) are adequately served by currently available 2697 prioritizing techniques. Such traffic can use up to 80% of the total 2698 bandwidth required. There is also a subset of non-time-critical 2699 traffic that must be reliable even though it is not time sensitive. 2701 In this use case the primary need for deterministic networking is to 2702 provide end-to-end delivery of M2M messages within specific timing 2703 constraints, for example in closed loop automation control. Today 2704 this level of determinism is provided by proprietary networking 2705 technologies. In addition, standard networking technologies are used 2706 to connect the local network to remote industrial automation sites, 2707 e.g. over an enterprise or metro network which also carries other 2708 types of traffic. Therefore, flows that should be forwarded with 2709 deterministic guarantees need to be sustained regardless of the 2710 amount of other flows in those networks. 2712 7.2. Industrial M2M Communication Today 2714 Today, proprietary networks fulfill the needed timing and 2715 availability for M2M networks. 2717 The network topologies used today by industrial automation are 2718 similar to those used by telecom networks: Daisy Chain, Ring, Hub and 2719 Spoke, and Comb (a subset of Daisy Chain). 2721 PLC-related control/data streams are transmitted periodically and 2722 carry either a pre-configured payload or a payload configured during 2723 runtime. 2725 Some industrial applications require time synchronization at the end 2726 nodes. For such time-coordinated PLCs, accuracy of 1 microsecond is 2727 required. Even in the case of "non-time-coordinated" PLCs time sync 2728 may be needed e.g. for timestamping of sensor data. 2730 Industrial network scenarios require advanced security solutions. 2731 Many of the current industrial production networks are physically 2732 separated. Preventing critical flows from be leaked outside a domain 2733 is handled today by filtering policies that are typically enforced in 2734 firewalls. 2736 7.2.1. Transport Parameters 2738 The Cycle Time defines the frequency of message(s) between industrial 2739 actors. The Cycle Time is application dependent, in the range of 1ms 2740 - 100ms for critical control/data streams. 2742 Because industrial applications assume deterministic transport for 2743 critical Control-Data-Stream parameters (instead of defining latency 2744 and delay variation parameters) it is sufficient to fulfill the upper 2745 bound of latency (maximum latency). The underlying networking 2746 infrastructure must ensure a maximum end-to-end delivery time of 2747 messages in the range of 100 microseconds to 50 milliseconds 2748 depending on the control loop application. 2750 The bandwidth requirements of control/data streams are usually 2751 calculated directly from the bytes-per-cycle parameter of the control 2752 loop. For PLC-to-PLC communication one can expect 2 - 32 streams 2753 with packet size in the range of 100 - 700 bytes. For S-PLC to PLCs 2754 the number of streams is higher - up to 256 streams. Usually no more 2755 than 20% of available bandwidth is used for critical control/data 2756 streams. In today's networks 1Gbps links are commonly used. 2758 Most PLC control loops are rather tolerant of packet loss, however 2759 critical control/data streams accept no more than 1 packet loss per 2760 consecutive communication cycle (i.e. if a packet gets lost in cycle 2761 "n", then the next cycle ("n+1") must be lossless). After two or 2762 more consecutive packet losses the network may be considered to be 2763 "down" by the Application. 2765 As network downtime may impact the whole production system the 2766 required network availability is rather high (99,999%). 2768 Based on the above parameters we expect that some form of redundancy 2769 will be required for M2M communications, however any individual 2770 solution depends on several parameters including cycle time, delivery 2771 time, etc. 2773 7.2.2. Stream Creation and Destruction 2775 In an industrial environment, critical control/data streams are 2776 created rather infrequently, on the order of ~10 times per day / week 2777 / month. Most of these critical control/data streams get created at 2778 machine startup, however flexibility is also needed during runtime, 2779 for example when adding or removing a machine. Going forward as 2780 production systems become more flexible, we expect a significant 2781 increase in the rate at which streams are created, changed and 2782 destroyed. 2784 7.3. Industrial M2M Future 2786 We would like to see a converged IP-standards-based network with 2787 deterministic properties that can satisfy the timing, security and 2788 reliability constraints described above. Today's proprietary 2789 networks could then be interfaced to such a network via gateways or, 2790 in the case of new installations, devices could be connected directly 2791 to the converged network. 2793 For this use case we expect time synchronization accuracy on the 2794 order of 1us. 2796 7.4. Industrial M2M Asks 2798 o Converged IP-based network 2800 o Deterministic behavior (bounded latency and jitter ) 2802 o High availability (presumably through redundancy) (99.999 %) 2804 o Low message delivery time (100us - 50ms) 2806 o Low packet loss (burstless, 0.1-1 %) 2808 o Security (e.g. prevent critical flows from being leaked between 2809 physically separated networks) 2811 8. Mining Industry 2813 8.1. Use Case Description 2815 The mining industry is highly dependent on networks to monitor and 2816 control their systems both in open-pit and underground extraction, 2817 transport and refining processes. In order to reduce risks and 2818 increase operational efficiency in mining operations, a number of 2819 processes have migrated the operators from the extraction site to 2820 remote control and monitoring. 2822 In the case of open pit mining, autonomous trucks are used to 2823 transport the raw materials from the open pit to the refining factory 2824 where the final product (e.g. Copper) is obtained. Although the 2825 operation is autonomous, the tracks are remotely monitored from a 2826 central facility. 2828 In pit mines, the monitoring of the tailings or mine dumps is 2829 critical in order to avoid any environmental pollution. In the past, 2830 monitoring has been conducted through manual inspection of pre- 2831 installed dataloggers. Cabling is not usually exploited in such 2832 scenarios due to the cost and complex deployment requirements. 2833 Currently, wireless technologies are being employed to monitor these 2834 cases permanently. Slopes are also monitored in order to anticipate 2835 possible mine collapse. Due to the unstable terrain, cable 2836 maintenance is costly and complex and hence wireless technologies are 2837 employed. 2839 In the underground monitoring case, autonomous vehicles with 2840 extraction tools travel autonomously through the tunnels, but their 2841 operational tasks (such as excavation, stone breaking and transport) 2842 are controlled remotely from a central facility. This generates 2843 video and feedback upstream traffic plus downstream actuator control 2844 traffic. 2846 8.2. Mining Industry Today 2848 Currently the mining industry uses a packet switched architecture 2849 supported by high speed ethernet. However in order to achieve the 2850 delay and packet loss requirements the network bandwidth is 2851 overestimated, thus providing very low efficiency in terms of 2852 resource usage. 2854 QoS is implemented at the Routers to separate video, management, 2855 monitoring and process control traffic for each stream. 2857 Since mobility is involved in this process, the connection between 2858 the backbone and the mobile devices (e.g. trucks, trains and 2859 excavators) is solved using a wireless link. These links are based 2860 on 802.11 for open-pit mining and leaky feeder for underground 2861 mining. 2863 Lately in pit mines the use of LPWAN technologies has been extended: 2864 Tailings, slopes and mine dumps are monitored by battery-powered 2865 dataloggers that make use of robust long range radio technologies. 2866 Reliability is usually ensured through retransmissions at L2. 2867 Gateways or concentrators act as bridges forwarding the data to the 2868 backbone ethernet network. Deterministic requirements are biased 2869 towards reliability rather than latency as events are slowly 2870 triggered or can be anticipated in advance. 2872 At the mineral processing stage, conveyor belts and refining 2873 processes are controlled by a SCADA system, which provides the in- 2874 factory delay-constrained networking requirements. 2876 Voice communications are currently served by a redundant trunking 2877 infrastructure, independent from current data networks. 2879 8.3. Mining Industry Future 2881 Mining operations and management are currently converging towards a 2882 combination of autonomous operation and teleoperation of transport 2883 and extraction machines. This means that video, audio, monitoring 2884 and process control traffic will increase dramatically. Ideally, all 2885 activities on the mine will rely on network infrastructure. 2887 Wireless for open-pit mining is already a reality with LPWAN 2888 technologies and it is expected to evolve to more advanced LPWAN 2889 technologies such as those based on LTE to increase last hop 2890 reliability or novel LPWAN flavours with deterministic access. 2892 One area in which DetNet can improve this use case is in the wired 2893 networks that make up the "backbone network" of the system, which 2894 connect together many wireless access points (APs). The mobile 2895 machines (which are connected to the network via wireless) transition 2896 from one AP to the next as they move about. A deterministic, 2897 reliable, low latency backbone can enable these transitions to be 2898 more reliable. 2900 Connections which extend all the way from the base stations to the 2901 machinery via a mix of wired and wireless hops would also be 2902 beneficial, for example to improve remote control responsiveness of 2903 digging machines. However to guarantee deterministic performance of 2904 a DetNet, the end-to-end underlying network must be deterministic. 2905 Thus for this use case if a deterministic wireless transport is 2906 integrated with a wire-based DetNet network, it could create the 2907 desired wired plus wireless end-to-end deterministic network. 2909 8.4. Mining Industry Asks 2911 o Improved bandwidth efficiency 2913 o Very low delay to enable machine teleoperation 2915 o Dedicated bandwidth usage for high resolution video streams 2917 o Predictable delay to enable realtime monitoring 2919 o Potential to construct a unified DetNet network over a combination 2920 of wired and deterministic wireless links 2922 9. Private Blockchain 2924 9.1. Use Case Description 2926 Blockchain was created with bitcoin, as a 'public' blockchain on the 2927 open Internet, however blockchain has also spread far beyond its 2928 original host into various industries such as smart manufacturing, 2929 logistics, security, legal rights and others. In these industries 2930 blockchain runs in designated and carefully managed network in which 2931 deterministic networking requirements could be addressed by Detnet. 2932 Such implementations are referred to as 'private' blockchain. 2934 The sole distinction between public and private blockchain is related 2935 to who is allowed to participate in the network, execute the 2936 consensus protocol and maintain the shared ledger. 2938 Today's networks treat the traffic from blockchain on a best-effort 2939 basis, but blockchain operation could be made much more efficient if 2940 deterministic networking service were available to minimize latency 2941 and packet loss in the network. 2943 9.1.1. Blockchain Operation 2945 A 'block' runs as a container of a batch of primary items such as 2946 transactions, property records etc. The blocks are chained in such a 2947 way that the hash of the previous block works as the pointer header 2948 of the new block, where confirmation of each block requires a 2949 consensus mechanism. When an item arrives at a blockchain node, the 2950 latter broadcasts this item to the rest of nodes which receive and 2951 verify it and put it in the ongoing block. Block confirmation 2952 process begins as the amount of items reaches the predefined block 2953 capacity, and the node broadcasts its proved block to the rest of 2954 nodes to be verified and chained. 2956 9.1.2. Blockchain Network Architecture 2958 Blockchain node communication and coordination is achieved mainly 2959 through frequent point to multi-point communication, however 2960 persistent point-to-point connections are used to transport both the 2961 items and the blocks to the other nodes. 2963 When a node initiates, it first requests the other nodes' address 2964 from a specific entity such as DNS, then it creates persistent 2965 connections each of with other nodes. If node A confirms an item, it 2966 sends the item to the other nodes via the persistent connections. 2968 As a new block in a node completes and gets proved among the nodes, 2969 it starts propagating this block towards its neighbor nodes. Assume 2970 node A receives a block, it sends invite message after verification 2971 to its neighbor B, B checks if the designated block is available, it 2972 responds get message to A if it is unavailable, and A send the 2973 complete block to B. B repeats the process as A to start the next 2974 round of block propagation. 2976 The challenge of blockchain network operation is not overall data 2977 rates, since the volume from both block and item stays between 2978 hundreds of bytes to a couple of mega bytes per second, but is in 2979 transporting the blocks with minimum latency to maximize efficiency 2980 of the blockchain consensus process. 2982 9.1.3. Security Considerations 2984 Security is crucial to blockchain applications, and todayt blockchain 2985 addresses its security issues mainly at the application level, where 2986 cryptography as well as hash-based consensus play a leading role 2987 preventing both double-spending and malicious service attack. 2988 However, there is concern that in the proposed use case of a private 2989 blockchain network which is dependent on deterministic properties, 2990 the network could be vulnerable to delays and other specific attacks 2991 against determinism which could interrupt service. 2993 9.2. Private Blockchain Today 2995 Today private blockchain runs in L2 or L3 VPN, in general without 2996 guaranteed determinism. The industry players are starting to realize 2997 that improving determinism in their blockchain networks could improve 2998 the performance of their service, but as of today these goals are not 2999 being met. 3001 9.3. Private Blockchain Future 3003 Blockchain system performance can be greatly improved through 3004 deterministic networking service primarily because it would 3005 accelerate the consensus process. It would be valuable to be able to 3006 design a private blockchain network with the following properties: 3008 o Transport of point to multi-point traffic in a coordinated network 3009 architecture rather than at the application layer (which typically 3010 uses point-to-point connections) 3012 o Guaranteed transport latency 3014 o Reduced packet loss (to the point where packet retransmission- 3015 incurred delay would be negligible.) 3017 9.4. Private Blockchain Asks 3019 o Layer 2 and Layer 3 multicast of blockchain traffic 3021 o Item and block delivery with bounded, low latency and negligible 3022 packet loss 3024 o Coexistence in a single network of blockchain and IT traffic. 3026 o Ability to scale the network by distributing the centralized 3027 control of the network across multiple control entities. 3029 10. Network Slicing 3031 10.1. Use Case Description 3033 Network slicing divides one physical network infrastructure into 3034 multiple logical networks. Each slice, corresponding to a logical 3035 network, uses resources and network functions independently from each 3036 other. Network slicing provides flexibility of resource allocation 3037 and service quality customization. 3039 Future services will demand network performance with a wide variety 3040 of characteristics such as high data rate, low latency, low loss 3041 rate, security and many other parameters. Ideally every service 3042 would have its own physical network satisfying its particular 3043 performance requirements, however that would be prohibitively 3044 expensive. Network slicing can provide a customized slice for a 3045 single service, and multiple slices can share the same physical 3046 network. This method can optimize the performance for the service at 3047 lower cost, and the flexibility of setting up and release the slices 3048 also allows the user to allocate the network resources dynamically. 3050 Unlike other DetNet use cases, Network slicing is not a specific 3051 application with specific deterministic requirements; it is proposed 3052 as a new requirement for the future network, which is still in 3053 discussion, and DetNet is a candidate solution for it. 3055 10.2. Network Slicing Use Cases 3057 Network Slicing is a core feature of 5G defined in 3GPP, which is 3058 currently under development. A Network Slice in mobile network is a 3059 complete logical network including Radio Access Network (RAN) and 3060 Core Network (CN). It provides telecommunication services and 3061 network capabilities, which may vary (or not) from slice to slice. 3063 A 5G bearer network is a typical use case of network slicing, 3064 including 3 service scenarios: enhanced Mobile Broadband (eMBB), 3065 Ultra-Reliable and Low Latency Communications (URLLC), and massive 3066 Machine Type Communications (mMTC). Each of these are described 3067 below. 3069 10.2.1. Enhanced Mobile Broadband (eMBB) 3071 eMBB focuses on services characterized by high data rates, such as 3072 high definition (HD) videos, virtual reality (VR), augmented reality 3073 (AR), and fixed mobile convergence (FMC). 3075 10.2.2. Ultra-Reliable and Low Latency Communications (URLLC) 3077 URLLC focuses on latency-sensitive services, such as self-driving 3078 vehicles, remote surgery, or drone control. 3080 10.2.3. massive Machine Type Communications (mMTC) 3082 mMTC focuses on services that have high requirements for connection 3083 density, such as those typical for smart city and smart agriculture 3084 use cases. 3086 10.3. Using DetNet in Network Slicing 3088 One of the requirements discussed for network slicing is the "hard" 3089 separation of various users' deterministic performance. That is, it 3090 should be impossible for activity, lack of activity, or changes in 3091 activity of one or more users to have any appreciable effect on the 3092 deterministic performance parameters of any other users. Typical 3093 techniques used today, which share a physical network among users, do 3094 not offer this kind of insulation. DetNet can supply point-to-point 3095 or point-to-multipoint paths that offer bandwidth and latency 3096 guarantees to a user that cannot be affected by other users' data 3097 traffic. 3099 Thus DetNet is a powerful tool when latency and reliability are 3100 required in Network Slicing. However, DetNet cannot cover every 3101 Network Slicing use case, and there are some other problems to be 3102 solved. Firstly, DetNet is a point-to-point or point-to-multipoint 3103 technology while Network Slicing needs multi-point to multi-point 3104 guarantee. Second, the number of flows that can be carried by DetNet 3105 is limited by DetNet scalability. Flow aggregation and queuing 3106 management modification may help to fix the problem. More work and 3107 discussions are needed in these topics. 3109 10.4. Network Slicing Today and Future 3111 Network slicing can satisfy the requirements of a lot of future 3112 deployment scenario, but it is still a collection of ideas and 3113 analysis, without a specific technical solution. A lot of 3114 technologies, such as Flex-E, Segment Routing, and DetNet have 3115 potential to be used in Network Slicing. For more details please see 3116 IETF99 Network Slicing BOF session agenda and materials. 3118 10.5. Network Slicing Asks 3120 o Isolation from other flows through Queuing Management 3122 o Service Quality Customization and Guarantee 3124 o Security 3126 11. Use Case Common Themes 3128 This section summarizes the expected properties of a DetNet network, 3129 based on the use cases as described in this draft. 3131 11.1. Unified, standards-based network 3133 11.1.1. Extensions to Ethernet 3135 A DetNet network is not "a new kind of network" - it based on 3136 extensions to existing Ethernet standards, including elements of IEEE 3137 802.1 AVB/TSN and related standards. Presumably it will be possible 3138 to run DetNet over other underlying transports besides Ethernet, but 3139 Ethernet is explicitly supported. 3141 11.1.2. Centrally Administered 3143 In general a DetNet network is not expected to be "plug and play" - 3144 it is expected that there is some centralized network configuration 3145 and control system. Such a system may be in a single central 3146 location, or it maybe distributed across multiple control entities 3147 that function together as a unified control system for the network. 3148 However, the ability to "hot swap" components (e.g. due to 3149 malfunction) is similar enough to "plug and play" that this kind of 3150 behavior may be expected in DetNet networks, depending on the 3151 implementation. 3153 11.1.3. Standardized Data Flow Information Models 3155 Data Flow Information Models to be used with DetNet networks are to 3156 be specified by DetNet. 3158 11.1.4. L2 and L3 Integration 3160 A DetNet network is intended to integrate between Layer 2 (bridged) 3161 network(s) (e.g. AVB/TSN LAN) and Layer 3 (routed) network(s) (e.g. 3162 using IP-based protocols). One example of this is "making AVB/TSN- 3163 type deterministic performance available from Layer 3 applications, 3164 e.g. using RTP". Another example is "connecting two AVB/TSN LANs 3165 ("islands") together through a standard router". 3167 11.1.5. Guaranteed End-to-End Delivery 3169 Packets sent over DetNet are guaranteed not to be dropped by the 3170 network due to congestion. (Packets may however be dropped for 3171 intended reasons, e.g. per security measures). 3173 11.1.6. Replacement for Multiple Proprietary Deterministic Networks 3175 There are many proprietary non-interoperable deterministic Ethernet- 3176 based networks currently available; DetNet is intended to provide an 3177 open-standards-based alternative to such networks. 3179 11.1.7. Mix of Deterministic and Best-Effort Traffic 3181 DetNet is intended to support coexistance of time-sensitive 3182 operational (OT) traffic and information (IT) traffic on the same 3183 ("unified") network. 3185 11.1.8. Unused Reserved BW to be Available to Best Effort Traffic 3187 If bandwidth reservations are made for a stream but the associated 3188 bandwidth is not used at any point in time, that bandwidth is made 3189 available on the network for best-effort traffic. If the owner of 3190 the reserved stream then starts transmitting again, the bandwidth is 3191 no longer available for best-effort traffic, on a moment-to-moment 3192 basis. Note that such "temporarily available" bandwidth is not 3193 available for time-sensitive traffic, which must have its own 3194 reservation. 3196 11.1.9. Lower Cost, Multi-Vendor Solutions 3198 The DetNet network specifications are intended to enable an ecosystem 3199 in which multiple vendors can create interoperable products, thus 3200 promoting device diversity and potentially higher numbers of each 3201 device manufactured, promoting cost reduction and cost competition 3202 among vendors. The intent is that DetNet networks should be able to 3203 be created at lower cost and with greater diversity of available 3204 devices than existing proprietary networks. 3206 11.2. Scalable Size 3208 DetNet networks range in size from very small, e.g. inside a single 3209 industrial machine, to very large, for example a Utility Grid network 3210 spanning a whole country, and involving many "hops" over various 3211 kinds of links for example radio repeaters, microwave linkes, fiber 3212 optic links, etc.. However recall that the scope of DetNet is 3213 confined to networks that are centrally administered, and explicitly 3214 excludes unbounded decentralized networks such as the Internet. 3216 11.3. Scalable Timing Parameters and Accuracy 3218 11.3.1. Bounded Latency 3220 The DetNet Data Flow Information Model is expected to provide means 3221 to configure the network that include parameters for querying network 3222 path latency, requesting bounded latency for a given stream, 3223 requesting worst case maximum and/or minimum latency for a given path 3224 or stream, and so on. It is an expected case that the network may 3225 not be able to provide a given requested service level, and if so the 3226 network control system should reply that the requested services is 3227 not available (as opposed to accepting the parameter but then not 3228 delivering the desired behavior). 3230 11.3.2. Low Latency 3232 Applications may require "extremely low latency" however depending on 3233 the application these may mean very different latency values; for 3234 example "low latency" across a Utility grid network is on a different 3235 time scale than "low latency" in a motor control loop in a small 3236 machine. The intent is that the mechanisms for specifying desired 3237 latency include wide ranges, and that architecturally there is 3238 nothing to prevent arbirtrarily low latencies from being implemented 3239 in a given network. 3241 11.3.3. Symmetrical Path Delays 3243 Some applications would like to specify that the transit delay time 3244 values be equal for both the transmit and return paths. 3246 11.4. High Reliability and Availability 3248 Reliablity is of critical importance to many DetNet applications, in 3249 which consequences of failure can be extraordinarily high in terms of 3250 cost and even human life. DetNet based systems are expected to be 3251 implemented with essentially arbitrarily high availability (for 3252 example 99.9999% up time, or even 12 nines). The intent is that the 3253 DetNet designs should not make any assumptions about the level of 3254 reliability and availability that may be required of a given system, 3255 and should define parameters for communicating these kinds of metrics 3256 within the network. 3258 A strategy used by DetNet for providing such extraordinarily high 3259 levels of reliability is to provide redundant paths that can be 3260 seamlessly switched between, while maintaining the required 3261 performance of that system. 3263 11.5. Security 3265 Security is of critical importance to many DetNet applications. A 3266 DetNet network must be able to be made secure against devices 3267 failures, attackers, misbehaving devices, and so on. In a DetNet 3268 network the data traffic is expected to be be time-sensitive, thus in 3269 addition to arriving with the data content as intended, the data must 3270 also arrive at the expected time. This may present "new" security 3271 challenges to implementers, and must be addressed accordingly. There 3272 are other security implications, including (but not limited to) the 3273 change in attack surface presented by packet replication and 3274 elimination. 3276 11.6. Deterministic Flows 3278 Reserved bandwidth data flows must be isolated from each other and 3279 from best-effort traffic, so that even if the network is saturated 3280 with best-effort (and/or reserved bandwidth) traffic, the configured 3281 flows are not adversely affected. 3283 12. Use Cases Explicitly Out of Scope for DetNet 3285 This section contains use case text that has been determined to be 3286 outside of the scope of the present DetNet work. 3288 12.1. DetNet Scope Limitations 3290 The scope of DetNet is deliberately limited to specific use cases 3291 that are consistent with the WG charter, subject to the 3292 interpretation of the WG. At the time the DetNet Use Cases were 3293 solicited and provided by the authors the scope of DetNet was not 3294 clearly defined, and as that clarity has emerged, certain of the use 3295 cases have been determined to be outside the scope of the present 3296 DetNet work. Such text has been moved into this section to clarify 3297 that these use cases will not be supported by the DetNet work. 3299 The text in this section was moved here based on the following 3300 "exclusion" principles. Or, as an alternative to moving all such 3301 text to this section, some draft text has been modified in situ to 3302 reflect these same principles. 3304 The following principles have been established to clarify the scope 3305 of the present DetNet work. 3307 o The scope of network addressed by DetNet is limited to networks 3308 that can be centrally controlled, i.e. an "enterprise" aka 3309 "corporate" network. This explicitly excludes "the open 3310 Internet". 3312 o Maintaining synchronized time across a DetNet network is crucial 3313 to its operation, however DetNet assumes that time is to be 3314 maintained using other means, for example (but not limited to) 3315 Precision Time Protocol ([IEEE1588]). A use case may state the 3316 accuracy and reliability that it expects from the DetNet network 3317 as part of a whole system, however it is understood that such 3318 timing properties are not guaranteed by DetNet itself. It is 3319 currently an open question as to whether DetNet protocols will 3320 include a way for an application to communicate such timing 3321 expectations to the network, and if so whether they would be 3322 expected to materially affect the performance they would receive 3323 from the network as a result. 3325 12.2. Internet-based Applications 3327 12.2.1. Use Case Description 3329 There are many applications that communicate across the open Internet 3330 that could benefit from guaranteed delivery and bounded latency. The 3331 following are some representative examples. 3333 12.2.1.1. Media Content Delivery 3335 Media content delivery continues to be an important use of the 3336 Internet, yet users often experience poor quality audio and video due 3337 to the delay and jitter inherent in today's Internet. 3339 12.2.1.2. Online Gaming 3341 Online gaming is a significant part of the gaming market, however 3342 latency can degrade the end user experience. For example "First 3343 Person Shooter" (FPS) games are highly delay-sensitive. 3345 12.2.1.3. Virtual Reality 3347 Virtual reality (VR) has many commercial applications including real 3348 estate presentations, remote medical procedures, and so on. Low 3349 latency is critical to interacting with the virtual world because 3350 perceptual delays can cause motion sickness. 3352 12.2.2. Internet-Based Applications Today 3354 Internet service today is by definition "best effort", with no 3355 guarantees on delivery or bandwidth. 3357 12.2.3. Internet-Based Applications Future 3359 We imagine an Internet from which we will be able to play a video 3360 without glitches and play games without lag. 3362 For online gaming, the maximum round-trip delay can be 100ms and 3363 stricter for FPS gaming which can be 10-50ms. Transport delay is the 3364 dominate part with a 5-20ms budget. 3366 For VR, 1-10ms maximum delay is needed and total network budget is 3367 1-5ms if doing remote VR. 3369 Flow identification can be used for gaming and VR, i.e. it can 3370 recognize a critical flow and provide appropriate latency bounds. 3372 12.2.4. Internet-Based Applications Asks 3374 o Unified control and management protocols to handle time-critical 3375 data flow 3377 o Application-aware flow filtering mechanism to recognize the timing 3378 critical flow without doing 5-tuple matching 3380 o Unified control plane to provide low latency service on Layer-3 3381 without changing the data plane 3383 o OAM system and protocols which can help to provide E2E-delay 3384 sensitive service provisioning 3386 12.3. Pro Audio and Video - Digital Rights Management (DRM) 3388 This section was moved here because this is considered a Link layer 3389 topic, not direct responsibility of DetNet. 3391 Digital Rights Management (DRM) is very important to the audio and 3392 video industries. Any time protected content is introduced into a 3393 network there are DRM concerns that must be maintained (see 3394 [CONTENT_PROTECTION]). Many aspects of DRM are outside the scope of 3395 network technology, however there are cases when a secure link 3396 supporting authentication and encryption is required by content 3397 owners to carry their audio or video content when it is outside their 3398 own secure environment (for example see [DCI]). 3400 As an example, two techniques are Digital Transmission Content 3401 Protection (DTCP) and High-Bandwidth Digital Content Protection 3402 (HDCP). HDCP content is not approved for retransmission within any 3403 other type of DRM, while DTCP may be retransmitted under HDCP. 3404 Therefore if the source of a stream is outside of the network and it 3405 uses HDCP protection it is only allowed to be placed on the network 3406 with that same HDCP protection. 3408 12.4. Pro Audio and Video - Link Aggregation 3410 Note: The term "Link Aggregation" is used here as defined by the text 3411 in the following paragraph, i.e. not following a more common Network 3412 Industry definition. Current WG consensus is that this item won't be 3413 directly supported by the DetNet architecture, for example because it 3414 implies guarantee of in-order delivery of packets which conflicts 3415 with the core goal of achieving the lowest possible latency. 3417 For transmitting streams that require more bandwidth than a single 3418 link in the target network can support, link aggregation is a 3419 technique for combining (aggregating) the bandwidth available on 3420 multiple physical links to create a single logical link of the 3421 required bandwidth. However, if aggregation is to be used, the 3422 network controller (or equivalent) must be able to determine the 3423 maximum latency of any path through the aggregate link. 3425 13. Acknowledgments 3427 13.1. Pro Audio 3429 This section was derived from draft-gunther-detnet-proaudio-req-01. 3431 The editors would like to acknowledge the help of the following 3432 individuals and the companies they represent: 3434 Jeff Koftinoff, Meyer Sound 3436 Jouni Korhonen, Associate Technical Director, Broadcom 3438 Pascal Thubert, CTAO, Cisco 3440 Kieran Tyrrell, Sienda New Media Technologies GmbH 3442 13.2. Utility Telecom 3444 This section was derived from draft-wetterwald-detnet-utilities-reqs- 3445 02. 3447 Faramarz Maghsoodlou, Ph. D. IoT Connected Industries and Energy 3448 Practice Cisco 3450 Pascal Thubert, CTAO Cisco 3452 13.3. Building Automation Systems 3454 This section was derived from draft-bas-usecase-detnet-00. 3456 13.4. Wireless for Industrial 3458 This section was derived from draft-thubert-6tisch-4detnet-01. 3460 This specification derives from the 6TiSCH architecture, which is the 3461 result of multiple interactions, in particular during the 6TiSCH 3462 (bi)Weekly Interim call, relayed through the 6TiSCH mailing list at 3463 the IETF. 3465 The authors wish to thank: Kris Pister, Thomas Watteyne, Xavier 3466 Vilajosana, Qin Wang, Tom Phinney, Robert Assimiti, Michael 3467 Richardson, Zhuo Chen, Malisa Vucinic, Alfredo Grieco, Martin Turon, 3468 Dominique Barthel, Elvis Vogli, Guillaume Gaillard, Herman Storey, 3469 Maria Rita Palattella, Nicola Accettura, Patrick Wetterwald, Pouria 3470 Zand, Raghuram Sudhaakar, and Shitanshu Shah for their participation 3471 and various contributions. 3473 13.5. Cellular Radio 3475 This section was derived from draft-korhonen-detnet-telreq-00. 3477 13.6. Industrial M2M 3479 The authors would like to thank Feng Chen and Marcel Kiessling for 3480 their comments and suggestions. 3482 13.7. Internet Applications and CoMP 3484 This section was derived from draft-zha-detnet-use-case-00. 3486 This document has benefited from reviews, suggestions, comments and 3487 proposed text provided by the following members, listed in 3488 alphabetical order: Jing Huang, Junru Lin, Lehong Niu and Oilver 3489 Huang. 3491 13.8. Electrical Utilities 3493 The wind power generation use case has been extracted from the study 3494 of Wind Farms conducted within the 5GPPP Virtuwind Project. The 3495 project is funded by the European Union's Horizon 2020 research and 3496 innovation programme under grant agreement No 671648 (VirtuWind). 3498 13.9. Network Slicing 3500 This section was written by Xuesong Geng, who would like to 3501 acknowledge Norm Finn and Mach Chen for their useful comments. 3503 13.10. Mining 3505 This section was written by Diego Dujovne in conjunction with Xavier 3506 Vilasojana. 3508 13.11. Private Blockchain 3510 This section was written by Daniel Huang. 3512 14. Informative References 3514 [ACE] IETF, "Authentication and Authorization for Constrained 3515 Environments", 3516 . 3518 [Ahm14] Ahmed, M. and R. Kim, "Communication network architectures 3519 for smart-wind power farms.", Energies, p. 3900-3921. , 3520 June 2014. 3522 [bacnetip] 3523 ASHRAE, "Annex J to ANSI/ASHRAE 135-1995 - BACnet/IP", 3524 January 1999. 3526 [CCAMP] IETF, "Common Control and Measurement Plane", 3527 . 3529 [CoMP] NGMN Alliance, "RAN EVOLUTION PROJECT COMP EVALUATION AND 3530 ENHANCEMENT", NGMN Alliance NGMN_RANEV_D3_CoMP_Evaluation_ 3531 and_Enhancement_v2.0, March 2015, 3532 . 3535 [CONTENT_PROTECTION] 3536 Olsen, D., "1722a Content Protection", 2012, 3537 . 3540 [CPRI] CPRI Cooperation, "Common Public Radio Interface (CPRI); 3541 Interface Specification", CPRI Specification V6.1, July 3542 2014, . 3545 [CPRI-transp] 3546 CPRI TWG, "CPRI requirements for Ethernet Fronthaul", 3547 November 2015, 3548 . 3551 [DCI] Digital Cinema Initiatives, LLC, "DCI Specification, 3552 Version 1.2", 2012, . 3554 [DICE] IETF, "DTLS In Constrained Environments", 3555 . 3557 [EA12] Evans, P. and M. Annunziata, "Industrial Internet: Pushing 3558 the Boundaries of Minds and Machines", November 2012. 3560 [ESPN_DC2] 3561 Daley, D., "ESPN's DC2 Scales AVB Large", 2014, 3562 . 3565 [flnet] Japan Electrical Manufacturers Association, "JEMA 1479 - 3566 English Edition", September 2012. 3568 [Fronthaul] 3569 Chen, D. and T. Mustala, "Ethernet Fronthaul 3570 Considerations", IEEE 1904.3, February 2015, 3571 . 3574 [HART] www.hartcomm.org, "Highway Addressable remote Transducer, 3575 a group of specifications for industrial process and 3576 control devices administered by the HART Foundation". 3578 [I-D.finn-detnet-architecture] 3579 Finn, N. and P. Thubert, "Deterministic Networking 3580 Architecture", draft-finn-detnet-architecture-08 (work in 3581 progress), August 2016. 3583 [I-D.finn-detnet-problem-statement] 3584 Finn, N. and P. Thubert, "Deterministic Networking Problem 3585 Statement", draft-finn-detnet-problem-statement-05 (work 3586 in progress), March 2016. 3588 [I-D.ietf-6tisch-6top-interface] 3589 Wang, Q. and X. Vilajosana, "6TiSCH Operation Sublayer 3590 (6top) Interface", draft-ietf-6tisch-6top-interface-04 3591 (work in progress), July 2015. 3593 [I-D.ietf-6tisch-architecture] 3594 Thubert, P., "An Architecture for IPv6 over the TSCH mode 3595 of IEEE 802.15.4", draft-ietf-6tisch-architecture-12 (work 3596 in progress), August 2017. 3598 [I-D.ietf-6tisch-coap] 3599 Sudhaakar, R. and P. Zand, "6TiSCH Resource Management and 3600 Interaction using CoAP", draft-ietf-6tisch-coap-03 (work 3601 in progress), March 2015. 3603 [I-D.ietf-6tisch-terminology] 3604 Palattella, M., Thubert, P., Watteyne, T., and Q. Wang, 3605 "Terminology in IPv6 over the TSCH mode of IEEE 3606 802.15.4e", draft-ietf-6tisch-terminology-09 (work in 3607 progress), June 2017. 3609 [I-D.ietf-ipv6-multilink-subnets] 3610 Thaler, D. and C. Huitema, "Multi-link Subnet Support in 3611 IPv6", draft-ietf-ipv6-multilink-subnets-00 (work in 3612 progress), July 2002. 3614 [I-D.ietf-mpls-residence-time] 3615 Mirsky, G., Ruffini, S., Gray, E., Drake, J., Bryant, S., 3616 and S. Vainshtein, "Residence Time Measurement in MPLS 3617 network", draft-ietf-mpls-residence-time-15 (work in 3618 progress), March 2017. 3620 [I-D.ietf-roll-rpl-industrial-applicability] 3621 Phinney, T., Thubert, P., and R. Assimiti, "RPL 3622 applicability in industrial networks", draft-ietf-roll- 3623 rpl-industrial-applicability-02 (work in progress), 3624 October 2013. 3626 [I-D.ietf-tictoc-1588overmpls] 3627 Davari, S., Oren, A., Bhatia, M., Roberts, P., and L. 3628 Montini, "Transporting Timing messages over MPLS 3629 Networks", draft-ietf-tictoc-1588overmpls-07 (work in 3630 progress), October 2015. 3632 [I-D.kh-spring-ip-ran-use-case] 3633 Khasnabish, B., hu, f., and L. Contreras, "Segment Routing 3634 in IP RAN use case", draft-kh-spring-ip-ran-use-case-02 3635 (work in progress), November 2014. 3637 [I-D.svshah-tsvwg-deterministic-forwarding] 3638 Shah, S. and P. Thubert, "Deterministic Forwarding PHB", 3639 draft-svshah-tsvwg-deterministic-forwarding-04 (work in 3640 progress), August 2015. 3642 [I-D.thubert-6lowpan-backbone-router] 3643 Thubert, P., "6LoWPAN Backbone Router", draft-thubert- 3644 6lowpan-backbone-router-03 (work in progress), February 3645 2013. 3647 [I-D.wang-6tisch-6top-sublayer] 3648 Wang, Q. and X. Vilajosana, "6TiSCH Operation Sublayer 3649 (6top)", draft-wang-6tisch-6top-sublayer-04 (work in 3650 progress), November 2015. 3652 [IEC-60870-5-104] 3653 International Electrotechnical Commission, "International 3654 Standard IEC 60870-5-104: Network access for IEC 3655 60870-5-101 using standard transport profiles", June 2006. 3657 [IEC61400] 3658 "International standard 61400-25: Communications for 3659 monitoring and control of wind power plants", June 2013. 3661 [IEC61850-90-12] 3662 TC57 WG10, IEC., "IEC 61850-90-12 TR: Communication 3663 networks and systems for power utility automation - Part 3664 90-12: Wide area network engineering guidelines", 2015. 3666 [IEC62439-3:2012] 3667 TC65, IEC., "IEC 62439-3: Industrial communication 3668 networks - High availability automation networks - Part 3: 3669 Parallel Redundancy Protocol (PRP) and High-availability 3670 Seamless Redundancy (HSR)", 2012. 3672 [IEEE1588] 3673 IEEE, "IEEE Standard for a Precision Clock Synchronization 3674 Protocol for Networked Measurement and Control Systems", 3675 IEEE Std 1588-2008, 2008, 3676 . 3679 [IEEE1646] 3680 "Communication Delivery Time Performance Requirements for 3681 Electric Power Substation Automation", IEEE Standard 3682 1646-2004 , Apr 2004. 3684 [IEEE1722] 3685 IEEE, "1722-2011 - IEEE Standard for Layer 2 Transport 3686 Protocol for Time Sensitive Applications in a Bridged 3687 Local Area Network", IEEE Std 1722-2011, 2011, 3688 . 3691 [IEEE19043] 3692 IEEE Standards Association, "IEEE 1904.3 TF", IEEE 1904.3, 3693 2015, . 3695 [IEEE802.1TSNTG] 3696 IEEE Standards Association, "IEEE 802.1 Time-Sensitive 3697 Networks Task Group", March 2013, 3698 . 3700 [IEEE802154] 3701 IEEE standard for Information Technology, "IEEE std. 3702 802.15.4, Part. 15.4: Wireless Medium Access Control (MAC) 3703 and Physical Layer (PHY) Specifications for Low-Rate 3704 Wireless Personal Area Networks". 3706 [IEEE802154e] 3707 IEEE standard for Information Technology, "IEEE standard 3708 for Information Technology, IEEE std. 802.15.4, Part. 3709 15.4: Wireless Medium Access Control (MAC) and Physical 3710 Layer (PHY) Specifications for Low-Rate Wireless Personal 3711 Area Networks, June 2011 as amended by IEEE std. 3712 802.15.4e, Part. 15.4: Low-Rate Wireless Personal Area 3713 Networks (LR-WPANs) Amendment 1: MAC sublayer", April 3714 2012. 3716 [IEEE8021AS] 3717 IEEE, "Timing and Synchronizations (IEEE 802.1AS-2011)", 3718 IEEE 802.1AS-2001, 2011, 3719 . 3722 [IEEE8021CM] 3723 Farkas, J., "Time-Sensitive Networking for Fronthaul", 3724 Unapproved PAR, PAR for a New IEEE Standard; 3725 IEEE P802.1CM, April 2015, 3726 . 3729 [IEEE8021TSN] 3730 IEEE 802.1, "The charter of the TG is to provide the 3731 specifications that will allow time-synchronized low 3732 latency streaming services through 802 networks.", 2016, 3733 . 3735 [IETFDetNet] 3736 IETF, "Charter for IETF DetNet Working Group", 2015, 3737 . 3739 [ISA100] ISA/ANSI, "ISA100, Wireless Systems for Automation", 3740 . 3742 [ISA100.11a] 3743 ISA/ANSI, "Wireless Systems for Industrial Automation: 3744 Process Control and Related Applications - ISA100.11a-2011 3745 - IEC 62734", 2011, . 3748 [ISO7240-16] 3749 ISO, "ISO 7240-16:2007 Fire detection and alarm systems -- 3750 Part 16: Sound system control and indicating equipment", 3751 2007, . 3754 [knx] KNX Association, "ISO/IEC 14543-3 - KNX", November 2006. 3756 [lontalk] ECHELON, "LonTalk(R) Protocol Specification Version 3.0", 3757 1994. 3759 [LTE-Latency] 3760 Johnston, S., "LTE Latency: How does it compare to other 3761 technologies", March 2014, 3762 . 3765 [MEF] MEF, "Mobile Backhaul Phase 2 Amendment 1 -- Small Cells", 3766 MEF 22.1.1, July 2014, 3767 . 3770 [METIS] METIS, "Scenarios, requirements and KPIs for 5G mobile and 3771 wireless system", ICT-317669-METIS/D1.1 ICT- 3772 317669-METIS/D1.1, April 2013, . 3775 [modbus] Modbus Organization, "MODBUS APPLICATION PROTOCOL 3776 SPECIFICATION V1.1b", December 2006. 3778 [MODBUS] Modbus Organization, Inc., "MODBUS Application Protocol 3779 Specification", Apr 2012. 3781 [net5G] Ericsson, "5G Radio Access, Challenges for 2020 and 3782 Beyond", Ericsson white paper wp-5g, June 2013, 3783 . 3785 [NGMN] NGMN Alliance, "5G White Paper", NGMN 5G White Paper v1.0, 3786 February 2015, . 3789 [NGMN-fronth] 3790 NGMN Alliance, "Fronthaul Requirements for C-RAN", March 3791 2015, . 3794 [OPCXML] OPC Foundation, "OPC XML-Data Access Specification", Dec 3795 2004. 3797 [PCE] IETF, "Path Computation Element", 3798 . 3800 [profibus] 3801 IEC, "IEC 61158 Type 3 - Profibus DP", January 2001. 3803 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3804 Requirement Levels", BCP 14, RFC 2119, 3805 DOI 10.17487/RFC2119, March 1997, 3806 . 3808 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3809 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 3810 December 1998, . 3812 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3813 "Definition of the Differentiated Services Field (DS 3814 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3815 DOI 10.17487/RFC2474, December 1998, 3816 . 3818 [RFC3031] Rosen, E., Viswanathan, A., and R. Callon, "Multiprotocol 3819 Label Switching Architecture", RFC 3031, 3820 DOI 10.17487/RFC3031, January 2001, 3821 . 3823 [RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V., 3824 and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP 3825 Tunnels", RFC 3209, DOI 10.17487/RFC3209, December 2001, 3826 . 3828 [RFC3393] Demichelis, C. and P. Chimento, "IP Packet Delay Variation 3829 Metric for IP Performance Metrics (IPPM)", RFC 3393, 3830 DOI 10.17487/RFC3393, November 2002, 3831 . 3833 [RFC3411] Harrington, D., Presuhn, R., and B. Wijnen, "An 3834 Architecture for Describing Simple Network Management 3835 Protocol (SNMP) Management Frameworks", STD 62, RFC 3411, 3836 DOI 10.17487/RFC3411, December 2002, 3837 . 3839 [RFC3444] Pras, A. and J. Schoenwaelder, "On the Difference between 3840 Information Models and Data Models", RFC 3444, 3841 DOI 10.17487/RFC3444, January 2003, 3842 . 3844 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3845 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3846 . 3848 [RFC3985] Bryant, S., Ed. and P. Pate, Ed., "Pseudo Wire Emulation 3849 Edge-to-Edge (PWE3) Architecture", RFC 3985, 3850 DOI 10.17487/RFC3985, March 2005, 3851 . 3853 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3854 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3855 2006, . 3857 [RFC4553] Vainshtein, A., Ed. and YJ. Stein, Ed., "Structure- 3858 Agnostic Time Division Multiplexing (TDM) over Packet 3859 (SAToP)", RFC 4553, DOI 10.17487/RFC4553, June 2006, 3860 . 3862 [RFC4903] Thaler, D., "Multi-Link Subnet Issues", RFC 4903, 3863 DOI 10.17487/RFC4903, June 2007, 3864 . 3866 [RFC4919] Kushalnagar, N., Montenegro, G., and C. Schumacher, "IPv6 3867 over Low-Power Wireless Personal Area Networks (6LoWPANs): 3868 Overview, Assumptions, Problem Statement, and Goals", 3869 RFC 4919, DOI 10.17487/RFC4919, August 2007, 3870 . 3872 [RFC5086] Vainshtein, A., Ed., Sasson, I., Metz, E., Frost, T., and 3873 P. Pate, "Structure-Aware Time Division Multiplexed (TDM) 3874 Circuit Emulation Service over Packet Switched Network 3875 (CESoPSN)", RFC 5086, DOI 10.17487/RFC5086, December 2007, 3876 . 3878 [RFC5087] Stein, Y(J)., Shashoua, R., Insler, R., and M. Anavi, 3879 "Time Division Multiplexing over IP (TDMoIP)", RFC 5087, 3880 DOI 10.17487/RFC5087, December 2007, 3881 . 3883 [RFC6282] Hui, J., Ed. and P. Thubert, "Compression Format for IPv6 3884 Datagrams over IEEE 802.15.4-Based Networks", RFC 6282, 3885 DOI 10.17487/RFC6282, September 2011, 3886 . 3888 [RFC6550] Winter, T., Ed., Thubert, P., Ed., Brandt, A., Hui, J., 3889 Kelsey, R., Levis, P., Pister, K., Struik, R., Vasseur, 3890 JP., and R. Alexander, "RPL: IPv6 Routing Protocol for 3891 Low-Power and Lossy Networks", RFC 6550, 3892 DOI 10.17487/RFC6550, March 2012, 3893 . 3895 [RFC6551] Vasseur, JP., Ed., Kim, M., Ed., Pister, K., Dejean, N., 3896 and D. Barthel, "Routing Metrics Used for Path Calculation 3897 in Low-Power and Lossy Networks", RFC 6551, 3898 DOI 10.17487/RFC6551, March 2012, 3899 . 3901 [RFC6775] Shelby, Z., Ed., Chakrabarti, S., Nordmark, E., and C. 3902 Bormann, "Neighbor Discovery Optimization for IPv6 over 3903 Low-Power Wireless Personal Area Networks (6LoWPANs)", 3904 RFC 6775, DOI 10.17487/RFC6775, November 2012, 3905 . 3907 [RFC7554] Watteyne, T., Ed., Palattella, M., and L. Grieco, "Using 3908 IEEE 802.15.4e Time-Slotted Channel Hopping (TSCH) in the 3909 Internet of Things (IoT): Problem Statement", RFC 7554, 3910 DOI 10.17487/RFC7554, May 2015, 3911 . 3913 [Spe09] Sperotto, A., Sadre, R., Vliet, F., and A. Pras, "A First 3914 Look into SCADA Network Traffic", IP Operations and 3915 Management, p. 518-521. , June 2009. 3917 [SRP_LATENCY] 3918 Gunther, C., "Specifying SRP Latency", 2014, 3919 . 3922 [STUDIO_IP] 3923 Mace, G., "IP Networked Studio Infrastructure for 3924 Synchronized & Real-Time Multimedia Transmissions", 2007, 3925 . 3928 [SyncE] ITU-T, "G.8261 : Timing and synchronization aspects in 3929 packet networks", Recommendation G.8261, August 2013, 3930 . 3932 [TEAS] IETF, "Traffic Engineering Architecture and Signaling", 3933 . 3935 [TS23401] 3GPP, "General Packet Radio Service (GPRS) enhancements 3936 for Evolved Universal Terrestrial Radio Access Network 3937 (E-UTRAN) access", 3GPP TS 23.401 10.10.0, March 2013. 3939 [TS25104] 3GPP, "Base Station (BS) radio transmission and reception 3940 (FDD)", 3GPP TS 25.104 3.14.0, March 2007. 3942 [TS36104] 3GPP, "Evolved Universal Terrestrial Radio Access 3943 (E-UTRA); Base Station (BS) radio transmission and 3944 reception", 3GPP TS 36.104 10.11.0, July 2013. 3946 [TS36133] 3GPP, "Evolved Universal Terrestrial Radio Access 3947 (E-UTRA); Requirements for support of radio resource 3948 management", 3GPP TS 36.133 12.7.0, April 2015. 3950 [TS36211] 3GPP, "Evolved Universal Terrestrial Radio Access 3951 (E-UTRA); Physical channels and modulation", 3GPP 3952 TS 36.211 10.7.0, March 2013. 3954 [TS36300] 3GPP, "Evolved Universal Terrestrial Radio Access (E-UTRA) 3955 and Evolved Universal Terrestrial Radio Access Network 3956 (E-UTRAN); Overall description; Stage 2", 3GPP TS 36.300 3957 10.11.0, September 2013. 3959 [TSNTG] IEEE Standards Association, "IEEE 802.1 Time-Sensitive 3960 Networks Task Group", 2013, 3961 . 3963 [UHD-video] 3964 Holub, P., "Ultra-High Definition Videos and Their 3965 Applications over the Network", The 7th International 3966 Symposium on VICTORIES Project PetrHolub_presentation, 3967 October 2014, . 3970 [WirelessHART] 3971 www.hartcomm.org, "Industrial Communication Networks - 3972 Wireless Communication Network and Communication Profiles 3973 - WirelessHART - IEC 62591", 2010. 3975 Authors' Addresses 3977 Ethan Grossman (editor) 3978 Dolby Laboratories, Inc. 3979 1275 Market Street 3980 San Francisco, CA 94103 3981 USA 3983 Phone: +1 415 645 4726 3984 Email: ethan.grossman@dolby.com 3985 URI: http://www.dolby.com 3986 Craig Gunther 3987 Harman International 3988 10653 South River Front Parkway 3989 South Jordan, UT 84095 3990 USA 3992 Phone: +1 801 568-7675 3993 Email: craig.gunther@harman.com 3994 URI: http://www.harman.com 3996 Pascal Thubert 3997 Cisco Systems, Inc 3998 Building D 3999 45 Allee des Ormes - BP1200 4000 MOUGINS - Sophia Antipolis 06254 4001 FRANCE 4003 Phone: +33 497 23 26 34 4004 Email: pthubert@cisco.com 4006 Patrick Wetterwald 4007 Cisco Systems 4008 45 Allees des Ormes 4009 Mougins 06250 4010 FRANCE 4012 Phone: +33 4 97 23 26 36 4013 Email: pwetterw@cisco.com 4015 Jean Raymond 4016 Hydro-Quebec 4017 1500 University 4018 Montreal H3A3S7 4019 Canada 4021 Phone: +1 514 840 3000 4022 Email: raymond.jean@hydro.qc.ca 4023 Jouni Korhonen 4024 Broadcom Corporation 4025 3151 Zanker Road 4026 San Jose, CA 95134 4027 USA 4029 Email: jouni.nospam@gmail.com 4031 Yu Kaneko 4032 Toshiba 4033 1 Komukai-Toshiba-cho, Saiwai-ku, Kasasaki-shi 4034 Kanagawa, Japan 4036 Email: yu1.kaneko@toshiba.co.jp 4038 Subir Das 4039 Applied Communication Sciences 4040 150 Mount Airy Road, Basking Ridge 4041 New Jersey, 07920, USA 4043 Email: sdas@appcomsci.com 4045 Yiyong Zha 4046 Huawei Technologies 4048 Email: zhayiyong@huawei.com 4050 Balazs Varga 4051 Ericsson 4052 Konyves Kalman krt. 11/B 4053 Budapest 1097 4054 Hungary 4056 Email: balazs.a.varga@ericsson.com 4058 Janos Farkas 4059 Ericsson 4060 Konyves Kalman krt. 11/B 4061 Budapest 1097 4062 Hungary 4064 Email: janos.farkas@ericsson.com 4065 Franz-Josef Goetz 4066 Siemens 4067 Gleiwitzerstr. 555 4068 Nurnberg 90475 4069 Germany 4071 Email: franz-josef.goetz@siemens.com 4073 Juergen Schmitt 4074 Siemens 4075 Gleiwitzerstr. 555 4076 Nurnberg 90475 4077 Germany 4079 Email: juergen.jues.schmitt@siemens.com 4081 Xavier Vilajosana 4082 Worldsensing 4083 483 Arago 4084 Barcelona, Catalonia 08013 4085 Spain 4087 Email: xvilajosana@worldsensing.com 4089 Toktam Mahmoodi 4090 King's College London 4091 Strand, London WC2R 2LS 4092 London, London WC2R 2LS 4093 United Kingdom 4095 Email: toktam.mahmoodi@kcl.ac.uk 4097 Spiros Spirou 4098 Intracom Telecom 4099 19.7 km Markopoulou Ave. 4100 Peania, Attiki 19002 4101 Greece 4103 Email: spis@intracom-telecom.com 4104 Petra Vizarreta 4105 Technical University of Munich, TUM 4106 Maxvorstadt, ArcisstraBe 21 4107 Munich, Germany 80333 4108 Germany 4110 Email: petra.vizarreta@lkn.ei.tum.de 4112 Daniel Huang 4113 ZTE Corporation, Inc. 4114 No. 50 Software Avenue 4115 Nanjing, Jiangsu 210012 4116 P.R. China 4118 Email: huang.guangping@zte.com.cn 4120 Xuesong Geng 4121 Huawei Technologies 4123 Email: gengxuesong@huawei.com 4125 Diego Dujovne 4126 Universidad Diego Portales 4128 Email: diego.dujovne@mail.udp.cl 4130 Maik Seewald 4131 Cisco Systems 4133 Email: maseewal@cisco.com