idnits 2.17.1 draft-liu-coinrg-requirement-03.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 seems to have RFC 2119 boilerplate text. -- The document date (July 12, 2020) is 1377 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'I-D.kutscher-coinrg-dir' is defined on line 312, but no explicit reference was found in the text == Outdated reference: A later version (-02) exists of draft-kutscher-coinrg-dir-01 Summary: 0 errors (**), 0 flaws (~~), 4 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Computing in Network Research Group P. Liu 3 Internet-Draft L. Geng 4 Intended status: Informational China Mobile 5 Expires: January 13, 2021 July 12, 2020 7 Requirement of Computing in network 8 draft-liu-coinrg-requirement-03 10 Abstract 12 New technology such as IOT, edge computing,etc. propose the 13 requirement of computing in network, so the convergence of network 14 and computing has become a trend. It will bring some new directions 15 and areas to be considered, such as the relationship between network 16 and computing, the influence of integrating computing to the network, 17 and so on. 19 This document points out the requirements of computing in network 20 according to the development of new Industry, including the network 21 and computing requirements. 23 Requirements Language 25 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 26 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 27 document are to be interpreted as described in RFC 2119 [RFC2119]. 29 Status of This Memo 31 This Internet-Draft is submitted in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF). Note that other groups may also distribute 36 working documents as Internet-Drafts. The list of current Internet- 37 Drafts is at https://datatracker.ietf.org/drafts/current/. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 This Internet-Draft will expire on January 13, 2021. 46 Copyright Notice 48 Copyright (c) 2020 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (https://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 Table of Contents 63 1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 2 64 2. Requirements of Network . . . . . . . . . . . . . . . . . . . 3 65 2.1. Precision . . . . . . . . . . . . . . . . . . . . . . . . 3 66 2.2. Concurrent . . . . . . . . . . . . . . . . . . . . . . . 4 67 2.3. Addressing . . . . . . . . . . . . . . . . . . . . . . . 4 68 2.4. Information interaction . . . . . . . . . . . . . . . . . 4 69 3. Requirements of computing . . . . . . . . . . . . . . . . . . 5 70 3.1. Computing resource deployment . . . . . . . . . . . . . . 5 71 3.2. Computing resource discovery . . . . . . . . . . . . . . 5 72 3.3. Computing resource reservation . . . . . . . . . . . . . 5 73 3.4. Computing aware scheduling . . . . . . . . . . . . . . . 6 74 3.5. Computing resource OAM . . . . . . . . . . . . . . . . . 6 75 4. Requirements of management . . . . . . . . . . . . . . . . . 6 76 4.1. Cross domain management . . . . . . . . . . . . . . . . . 6 77 4.2. Joint optimisation . . . . . . . . . . . . . . . . . . . 6 78 4.3. Multi user access . . . . . . . . . . . . . . . . . . . . 7 79 5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . 7 80 6. Security Considerations . . . . . . . . . . . . . . . . . . . 7 81 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 82 8. Normative References . . . . . . . . . . . . . . . . . . . . 7 83 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 8 85 1. Overview 87 The new services' provider expects a user experience with lower 88 latency and high reliability, which put forwards immense challenges 89 to cloud computing and traditional network. Centralized computing 90 requires a long transmission distance of traffic flow, and the 91 existing network technology is to the best of its ability. Network 92 operators start to think about how to meet the higher needs of 93 service provider and users. Computing in the network may solve the 94 problems because it can provide a flexible network and computing 95 integration system. 97 To integrate the computing resource to the network, it need to find 98 suitable computing nodes to handle service's request, as well as a 99 forwarding path to them. How much computing resources will affect 100 the delay of service processing, which could also affect the whole 101 network latency. Just as the measurement of network performance has 102 one more dimension, it will interact and cooperate with others. So 103 there are some requirments for both network and computing. 105 2. Requirements of Network 107 The network requirements includes precision, concurrent, addressing 108 and information interaction. 110 2.1. Precision 112 Precision of the network refers to the deterministic of latency, 113 packet loss rate and perception of computing resources. 115 * Latency: The traditional network's best-effort forwarding mode can 116 no longer meet the demand of such services for network latency. The 117 deterministic latency brings forward a new measure latitude for 118 network, which changes from in-time to on-time. 120 * Packet loss rate: It is another factor to evaluate the precision of 121 the network. Utilizing the ubiquitous computing capability of the 122 network, network prediction and segment-by-segment path 123 retransmitting are realized based on AI, network transmission can be 124 optimized and service QoS can be ensured. 126 * Perception of computing resources: how to precisely obtain the 127 status of computing resource to meet the requirements of business 128 requests is also a challenge to the network. It considers the 129 network status and the performance status of computing resources can 130 be matched dynamicly. So the user experience, utilization rate of 131 computing resources and the network efficiency can be optimum. 133 For the latency and packet loss rate, some technologies such as time- 134 sensitive network TSN, deterministic network DetNet, etc., have 135 proposed corresponding technical means to provide network bearers 136 with deterministic latency(IEEE802.1Qbv, IEEE802.1Qbu) and packet 137 loss rate and guarantee the user's business experience. However, it 138 also needs to consider how to guarantee the service's end-to-end 139 latency, packet loss rate and resource utilization rate. 141 For the perception of computing resources, we may consider about the 142 OAM and telemetry to achieve it, however, the performance and 143 information collection strategies are issues that need attention. 145 2.2. Concurrent 147 There will be number of computing nodes deployed in the network, or 148 computing fucntions intergrated in the network device for network 149 computing. A serivce's computing request may distributed in several 150 computing nodes in order to response quickly to the client. So there 151 may be a lot of parallel computing task, whitch cause too much 152 connection among the nodes but consume less bandwidth. It will bring 153 great challenges to the concurrent network connection including how 154 to build and deploy these distributed computing nodes to ensure the 155 processing capability of the network , as well as the storage, call 156 of the database are worth studying. 158 2.3. Addressing 160 Traditional application-based addressing can not accurately grasp the 161 network performance in real time. The comprehensive performance of 162 addressing results based on application layer may not be the best. 163 It is always to find the consistent host's address and go through a 164 long distance internet, which results in poor business experience. 166 It need to find some new way to improve the addressing proccess. For 167 example, in the function based addressing, the application 168 deconstruction components on the server side are distributed on the 169 cloud platform, and the business logic in the server is transferred 170 to the client side. So the client only needs to care about the 171 computing function itself, not about the computing resources such as 172 server, virtual machine, container and so on. 174 2.4. Information interaction 176 The network needs to have the ability to sense application's 177 requirments and expose network and computing status. For example, 178 application can tell the network requirements including bandwidth, 179 latency and jitter, as well as the computing requirements, such as 180 CPU, storage and memory. The network also can have the capability to 181 be aware of the application's requirements. Thus it can effectively 182 support the network programming, which could meet the future business 183 requirements. 185 3. Requirements of computing 187 The computing requirements includes computing resource deployment, 188 discovery, reservation, scheduling and OAM. 190 3.1. Computing resource deployment 192 If some computing tasks in the network is planned to be implemented, 193 it needs to consider about what kinds of chips and where should them 194 be deployed. On the one hands, different kinds of computing require 195 different kinds of chips, such as CPU, GPU and memory chip. On the 196 other hands, those chips may be put into router, switch, server or 197 some dedicated machines, which are connected by the network. 199 There is an example about AI algorithm which might be discussed 200 before. The AI algorithm has several steps including training and 201 matching, and they also have different requirements of chips. In 202 network computing, those steps could be distributed in different 203 computing nodes. 205 3.2. Computing resource discovery 207 The network needs to have the ability to discover computing 208 resources. when the computing nodes are deployed in the network, it 209 need to be registered to the network management system, and the 210 information of computing resource or routing can update. In this 211 way, when there are computing tasks to be executed, the network can 212 reasonably allocate resources according to the needs of the 213 application. 215 3.3. Computing resource reservation 217 There might be serial distributed computing model of computing in the 218 network, and different resources need to be reserved for different 219 nodes. For example, AI algorithm now has a model of step-by-step 220 iteration at multiple nodes. The previous iteration will affect the 221 next calculation results, and the computing resources required for 222 each iteration are not the same. From the perspective of network 223 standard, we hope to regard computing resources as the dimensions to 224 measure network performance, such as the same bandwidth, path, etc., 225 while the traditional technologies of resource reservation have not 226 considered the reservation of computing resources, and have not 227 considered the differentiated resource reservation model. Therefore, 228 new protocol or extension of existing protocol is needed to meet the 229 requirement. 231 3.4. Computing aware scheduling 233 Computing in network needs a reasonable scheduling strategy, which 234 means computing aware scheduling. According to the business 235 requests, dynamicly computing power matching is carried out based on 236 network status and performance of computing resources to achieve 237 optimal user experience, optimal utilization of computing resources 238 and optimal network efficiency. In computing aware scheduling, 239 computing is seen as "link state" and the computing resource 240 information should be exposed. 242 3.5. Computing resource OAM 244 The ability of OAM can be used to continuously update the current 245 computing power resources, and perform some troubleshooting tasks. 246 However, OAM of computing resource is more complex than network. 247 Network monitoring is relatively simple, like bandwidth, latency, 248 jitter, while computing can be divided into many categories, 249 different application need different kinds of computing. So it need 250 to implement fine-grained OAM of computing resource. 252 4. Requirements of management 254 The management requirements includes cross domain management, joint 255 optimisation and multi user access. 257 4.1. Cross domain management 259 The computing in the network should ensure the end-to-end network 260 management to meet the needs of different network topology, 261 performance and function, which involves cross domain network 262 arrangement. In the process of network data transmission, different 263 services will forward in different ways or different network 264 protocols, and computing resources may be distributed in different 265 network domains. Effective cross domain management will enhance the 266 performance of network and computing. 268 4.2. Joint optimisation 270 As computing resources are integrated in the network, and may be used 271 as one of the measurement dimensions of network performance, joint 272 optimization is also a very important part. Network and computing 273 resources will affect each other, including performance, scheduling 274 and so on. So It need a good joint optimization scheduling strategy. 276 4.3. Multi user access 278 Many existing applications, such as games, remote video conferencing, 279 are usually multi--accessed and interacted by several users at the 280 same time. This brings about the problem of service consistency, 281 that is, users accessing to the same game or video need the 282 consistency of SLA, otherwise it will seriously affect the experience 283 of other users. Service consistency can be achieved through network 284 management or application layer control. 286 5. Conclusion 288 Based on the requirements of new business, this document puts forward 289 the requirements of computing in network, and gives some reference 290 technologies and use cases. Computing in network is a new direction, 291 some details need more in-depth discussion and research. 293 6. Security Considerations 295 Computing In network has brought the trend of network convergence in 296 different regions. For example, 5g network of operators can go deep 297 into the vertical industry user site to provide users with higher 298 quality network services, which will bring the convergence of 299 operator's network and user site network. Besides, industrial 300 Internet brings the trend of integration of industrial OT network and 301 IT network to further improve the production efficiency of the 302 industry. It need to ensure the security of the network, including 303 the mutual trust and non aggression of information among regions, 304 which may require further protection and detection measures. 306 7. IANA Considerations 308 TBD. 310 8. Normative References 312 [I-D.kutscher-coinrg-dir] 313 Kutscher, D., Karkkainen, T., and J. Ott, "Directions for 314 Computing in the Network", draft-kutscher-coinrg-dir-01 315 (work in progress), November 2019. 317 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 318 Requirement Levels", BCP 14, RFC 2119, 319 DOI 10.17487/RFC2119, March 1997, 320 . 322 Authors' Addresses 324 Peng Liu 325 China Mobile 326 Beijing 100053 327 China 329 Email: liupengyjy@chinamobile.com 331 Liang Geng 332 China Mobile 333 Beijing 100053 334 China 336 Email: gengliang@chinamobile.com