idnits 2.17.1 draft-li-ippm-ref-delay-measurement-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document doesn't use any RFC 2119 keywords, yet has text resembling RFC 2119 boilerplate text. -- The document date (26 July 2021) is 1002 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Possible downref: Non-RFC (?) normative reference: ref. 'IEEE.1588.2008' Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group Y. Li 3 Internet-Draft T. Sun 4 Intended status: Standards Track H. Yang 5 Expires: 27 January 2022 D. Chen 6 China Mobile 7 Y. Wang 8 Huawei 9 26 July 2021 11 One-way Delay Measurement Based on Reference Delay 12 draft-li-ippm-ref-delay-measurement-01 14 Abstract 16 The end-to-end network one-way delay is an important performance 17 metric in the 5G network. For realizing the accurate one-way delay 18 measurement, existing methods requires the end-to-end deployment of 19 accurate clock synchronization mechanism, such as PTP or GPS, which 20 results in relatively high deployment cost. Another method can 21 derive the one-way delay from the round-trip delay. In this case, 22 since the delay of the downlink and uplink of the 5G network may be 23 asymmetric, the measurement accuracy is relatively low. Hence, this 24 document introduces a method to measure the end-to-end network one- 25 way delay based on a reference delay guaranteed by deterministic 26 networking without clock synchronization. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on 27 January 2022. 45 Copyright Notice 47 Copyright (c) 2021 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 52 license-info) in effect on the date of publication of this document. 53 Please review these documents carefully, as they describe your rights 54 and restrictions with respect to this document. Code Components 55 extracted from this document must include Simplified BSD License text 56 as described in Section 4.e of the Trust Legal Provisions and are 57 provided without warranty as described in the Simplified BSD License. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 62 2. Conventions Used in This Document . . . . . . . . . . . . . . 4 63 2.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 4 64 2.2. Requirements Language . . . . . . . . . . . . . . . . . . 4 65 3. The method of One-way Delay Measurement Based on Reference 66 Delay . . . . . . . . . . . . . . . . . . . . . . . . . . 4 67 3.1. One-way Delay Measurement Method . . . . . . . . . . . . 5 68 3.2. Packet and Measurement Header Format . . . . . . . . . . 7 69 4. Acquisition of Reference Delay . . . . . . . . . . . . . . . 8 70 5. Security Considerations . . . . . . . . . . . . . . . . . . . 8 71 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 72 7. Normative References . . . . . . . . . . . . . . . . . . . . 8 73 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9 75 1. Introduction 77 With the gradual promotion of new-generation network technologies 78 (such as 5G networks) and their application in various industries, 79 SLA guarantees for network quality become more and more important. 80 For example, different 5G services have different requirements for 81 network performance indicators such as delay, jitter, packet loss, 82 and bandwidth. Among them, the 5G network delay is defined as end- 83 to-end one-way delay of the network. Real-time and accurate 84 measurement of the end-to-end one-way delay is very important for the 85 SLA guarantee of network services, and has become an urgent and 86 important requirement. 88 As shown in figure 1, 5G network HD video surveillance service is a 89 common scenario having requirement of end-to-end one-way delay 90 measurement. In this case, one end of the network is a high- 91 definition surveillance camera in the wireless access side, and the 92 other end of the network is a video server. The end-to-end one-way 93 delay from the surveillance camera to the video server is the sum of 94 T1, T2, T3 and T4, which is composed of delay in wireless access 95 network, optical transmission network, 5G core network, and IP data 96 network. 98 +--------+ +-------+ +-------+ +-------+ 99 +------+ |Wireless| |Optical| |5G Core| | IP | +------+ 100 |Camera+<->+ Access +<->+ Trans +<->+Network+<->+ Data +<->+Server| 101 +------+ |Network | |Network| | | |Network| +------+ 102 +--------+ +-------+ +-------+ +-------+ 104 |<---- T1 ---->|<--- T2 -->|<--- T3 -->|<--- T4 ---->| 106 Figure 1: Figure 1:A Scenario for End-to-end One-way Delay 108 The existing one-way delay measurement solutions are divided into two 109 types. One type of mechanism to calculate one-way delay is based on 110 the measurement of round-trip delay. However, for example, because 111 upstream traffic and downstream traffic do not share the same path in 112 5G network, the accuracy of the end-to-end one-way delay calculated 113 from the round-trip delay is low. Another type of mechanism is in- 114 band OAM with accurate network time synchronization mechanism , such 115 as NTP[RFC5905] or PTP[IEEE.1588.2008]. 117 The one-way delay measurement solution based on precise network time 118 synchronization requires the deployment of an end-to-end time 119 synchronization mechanism. The current time synchronization accuracy 120 based on the NTP protocol can only reach millisecond level, which 121 cannot fully meet the measurement accuracy requirements. The time 122 synchronization accuracy based on the GPS module or the PTP protocol 123 can meet the requirements. However, because many data centers are 124 actually located underground or in rooms without GPS signals, so GPS 125 clock information cannot be continuously obtained for time 126 synchronization. For time synchronization solutions based on the PTP 127 protocol, each device in the wireless access network, 5G transport 128 network, and 5G core network must support the PTP protocol, which is 129 unrealistic at the moment. So the one-way delay measurement solution 130 based on precise end-to-end time synchronization is expensive and 131 difficult to be deployed. 133 This document introduces a one-way delay measurement mechanism for 134 Deterministic Networking (DetNet) [RFC8655]. The one-way delay 135 measurement is based on a stable one-way delay of a reference DetNet 136 packet, named as reference delay, which is known in advance and has 137 extremely low jitter. We can use the reference delay provided by the 138 reference DetNet packet to derive the one-way delay of other common 139 service packets. 141 2. Conventions Used in This Document 143 2.1. Terminology 145 NTP Network Time Protocol 147 PTP Precision Time Protocol 149 SLA Service Level Agreement 151 2.2. Requirements Language 153 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 154 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 155 "OPTIONAL" in this document are to be interpreted as described in BCP 156 14[RFC2119][RFC8174] when, and only when, they appear in all 157 capitals, as shown here. 159 3. The method of One-way Delay Measurement Based on Reference Delay 161 The end-to-end one-way delay of a reference packet with a stable 162 delay in the network can be used as a reference delay, denoted as 163 Dref, which is known in advance and has extremely low jitter. This 164 section will describe in detail the end-to-end one-way delay 165 measurement method based on reference delay of the reference packet. 166 Assume that the end-to-end one-way delay from the sender to the 167 receiver is measured, as shown in figure 2. The intermediate network 168 devices other than the sender and receiver are hidden in the figure. 170 +------------+ +------------+ 171 | | Clock Offset | | 172 | Sender | +---------------> | Receiver | 173 | | | | 174 +------------+ +------------+ 176 Reference +-----+ Dref +-----+ 177 Packet: | Ts1 | +-------------------> | Tr1 | 178 +-----+ +-----+ 180 Target +-----+ Dtarget +-----+ 181 Packet: | Ts2 | +-------------------> | Tr2 | 182 +-----+ +-----+ 184 Figure 2: Figure 2:Topology of One-way Delay Measurement 186 3.1. One-way Delay Measurement Method 188 The measurement steps are shown in figure 3, which describe the 189 measurement steps at the sender side and receiver side respectively. 190 For the sender side, a reference packet is sent. In the first step, 191 the sender gets ready to send a reference packet; in the second step, 192 the sender marks an egress timestamp Ts1 for the reference packet; in 193 the third step, the sender encapsulates the egress timestamp of the 194 reference packet in the measurement header of the reference packet; 195 in the fourth step, the sender sends the reference packet. For the 196 target packet, the sender side procedures are the same,we omit it for 197 simplicity. The sending time of the target packet is according to 198 the traffic model of real applications. On the other hand, the 199 sender can send the reference packet according to a fixed frequency 200 or adjust the sending frequency according to the link usage rate, so 201 that the target packet can always find a nearby reference packet to 202 make sure that the sending time interval between the reference packet 203 and the target packet is small. 205 For the reference packet, the processing steps at the receiver are 206 shown in figure 3. In the first step, the reference packet arrives 207 at the receiver, and the receiver receives the reference packet; in 208 the second step, the receiver timestamps the reference packet at the 209 entrance, which is denoted as Tr1; in the third step, the receiver 210 decapsulates the measurement header of the reference packet to obtain 211 the sender side timestamp Ts1; in the fourth step, the receiver 212 records the timestamp information of Ts1 and Tr1; in the fifth step, 213 the receiver uses the source/destination pair obtained by 214 decapsulation in the third step as the search key, queries the 215 reference delay table and records the reference delay search result, 216 denoted as Dref. 218 For the target packet, the processing steps at the receiver are also 219 shown in figure 3. In the first step, the target packet arrives at 220 the receiver, and the receiver receives the target packet; in the 221 second step, the receiver timestamps the target packet at the 222 entrance, which is denoted as Tr2; in the third step, the receiver 223 decapsulates the measurement header of the target packet to obtain 224 the sender side timestamp Ts2; in the fourth step, the receiver 225 records the timestamp information of Ts2 and Tr2; in the fifth step, 226 the receiver calculates the target one-way delay, which we want to 227 measure, according to the recorded timestamp information Ts1, Ts2, 228 Tr1, Tr2 and reference delay information Dref. The target one-way 229 delay of the target packet is recorded as Dtarget. 231 Sender Side Procedures for both Reference and Target Packet: 233 +-------+ +------------+ +-------------+ +-------+ 234 |Sender | |Sender Side | |Sender Side | |Sending| 235 |Ready +-->+Timestamping+-->+Encapsulation+-->+ Packet| 236 | | | | | | | | 237 +-------+ +------------+ +-------------+ +-------+ 239 Receiver Side Procedures for Reference Packet: 241 +---------+ +-------------+ +-------------+ +---------+ +---------+ 242 |Reference| |Receiver Side| |Receiver Side| |Timestamp| |Query for| 243 |Packet +->+Timestamping +->+Decapsulation+->+Recorded +->+Reference| 244 |Arrival | | | | | | | | Delay | 245 +---------+ +-------------+ +-------------+ +---------+ +---------+ 247 Receiver Side Procedures for Target Packet: 249 +-------+ +-------------+ +-------------+ +---------+ +-----------+ 250 | Target| |Receiver Side| |Receiver Side| |Timestamp| | One-way | 251 | Packet+->+Timestamping +->+Decapsulation+->+Recorded +->+ Delay | 252 |Arrival| | | | | | | |Calculation| 253 +-------+ +-------------+ +-------------+ +---------+ +-----------+ 255 Figure 3: Figure 3: Measurement steps for Sender and Receiver 256 Respectively 258 Now we describe the fifth step of the receiver procedures for the 259 target packet in figure 3, that is, calculating the one-way delay 260 Dtarget of the target packet based on the recorded timestamp 261 information Ts1, Ts2, Tr1, Tr2 and the reference delay information 262 Dref. The calculation method is the core of this solution. For the 263 reference packet, leveraging the receiver timestamp minus the sender 264 timestamp, we can get Equation 1. 266 Equation 1: Tr1 - Ts1 = Dref + Offset1 268 where Offset1 is the time offset between the sender and the receiver 269 when the reference packet transmission occurs. Similarly, for the 270 target packet, we can get Equation 2 using the same method. 272 Equation 2: Tr2 - Ts2 = Dtarget + Offset2 274 where Offset2 is the time offset between the sender and the receiver 275 when the target packet transmission occurs. Assuming that the 276 sending time interval between the reference packet and the target 277 packet is very small, we can get that Offset1 = Offset2. By 278 (Equation 2 - Equation 1), we can get Equation 3. 280 Equation 3: Dtarget = (Tr2 + Ts1) - (Tr1 + Ts2) + Dref 282 So the one-way delay of the target packet can be calculated by 283 Equation 3. 285 3.2. Packet and Measurement Header Format 287 The sender encapsulates the timestamp information and sender-receiver 288 pair information in the measurement header of the sent packet, as 289 shown in figure 4. The position of measurement header is in the 290 option field of the TCP protocol header. The delay measurement 291 option format is defined in figure 5. The Length value is 8 octets, 292 which is in accordance with TCP option. The sender ID is one octet, 293 and the receiver ID is also one octet. The sender side timestamp is 294 4 octets, which can store accurate timestamp information. 296 0 1 2 3 297 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 298 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 299 | | 300 | Ethernet header (14 octets) | 301 | | 302 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 303 | | 304 | IP header (20 octets) | 305 | | 306 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 307 | | 308 | TCP header (20 octets) | 309 | | 310 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 311 | TCP Delay Measurement Option (8 octets) | 312 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 313 | Data | 314 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 316 Figure 4: Figure 4: Format of Reference or Target Packet 318 0 1 2 3 319 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 320 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 321 | Kind=TBA | Length | Sender ID | Receiver ID | 322 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 323 | Sender Side Timestamp (4 octets) | 324 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 326 Figure 5: Figure 5: TCP Delay Measurement Option Format 328 4. Acquisition of Reference Delay 330 The end-to-end one-way delay includes three parts, namely the 331 transmission delay, the internal processing delay of the network 332 devices, and the internal queueing delay of the network devices. 333 Among them, fixed parts of the delay include transmission delay and 334 internal processing delay. The transmission delay is related to 335 transmission distance and transmission media. For example, in 336 optical fiber, it is about 5ns per meter. With transmission path and 337 media determined, it is basically a fixed value. The internal 338 processing delay of a network device includes processing delay of the 339 device's internal pipeline or processor and serial-to-parallel 340 conversion delay of the interface, which is related to in/out port 341 rate of the device, message length and forwarding behavior. The 342 magnitude of the internal processing delay is at microsecond level, 343 and it is basically a fixed value related to the chip design 344 specifications of a particular network device. Variable part of the 345 delay is the internal queueing delay. The queueing delay of the 346 device internal buffer is related to the queue depth, queue 347 scheduling algorithm, message priority and message length. For each 348 device along the end-to-end path, the queueing delay can reach 349 microsecond or even millisecond level, depending on values of the 350 above parameters and network congestion state. 352 With the continuous development of networking technologies and 353 application requirements, a series of new network technologies have 354 emerged which can guarantee bounded end-to-end delay and ultra small 355 jitter. For example, deterministic network[RFC8655], by leveraging 356 novel scheduling algorithms and packet priority settings, can 357 stabilize queuing delay of network device on the end-to-end path. As 358 a result, the end-to-end one-way delay is extremely low and bounded. 359 So packets transmitted by a deterministic network with delay 360 guarantee can be used as reference packets, and their end-to-end one- 361 way delay can be used as reference delays. The acquisition method of 362 reference delay is not limited to the above method based on 363 deterministic network technology. 365 5. Security Considerations 367 TBD. 369 6. IANA Considerations 371 This document requests IANA to assign a Kind Number in TCP Option to 372 indicate TCP Delay Measurement option. 374 7. Normative References 376 [IEEE.1588.2008] 377 IEEE, "IEEE Standard for a Precision Clock Synchronization 378 Protocol for Networked Measurement and Control Systems", 379 July 2008. 381 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 382 Requirement Levels", BCP 14, RFC 2119, 383 DOI 10.17487/RFC2119, March 1997, 384 . 386 [RFC5905] Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, 387 "Network Time Protocol Version 4: Protocol and Algorithms 388 Specification", RFC 5905, DOI 10.17487/RFC5905, June 2010, 389 . 391 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 392 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 393 May 2017, . 395 [RFC8655] Finn, N., Thubert, P., Varga, B., and J. Farkas, 396 "Deterministic Networking Architecture", RFC 8655, 397 DOI 10.17487/RFC8655, October 2019, 398 . 400 Authors' Addresses 402 Yang Li 403 China Mobile 404 Beijing 405 100053 406 China 408 Email: liyangzn@chinamobile.com 410 Tao Sun 411 China Mobile 412 Beijing 413 100053 414 China 416 Email: suntao@chinamobile.com 417 Hongwei Yang 418 China Mobile 419 Beijing 420 100053 421 China 423 Email: yanghongwei@chinamobile.com 425 Danyang Chen 426 China Mobile 427 Beijing 428 100053 429 China 431 Email: chendanyang@chinamobile.com 433 Yali Wang 434 Huawei 435 156 Beiqing Rd., Haidian District 436 Beijing 437 China 439 Email: wangyali11@huawei.com