idnits 2.17.1 draft-hongcs-chit-mec-00.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: ---------------------------------------------------------------------------- == There are 2 instances of lines with non-ascii characters in the document. == It seems as if not all pages are separated by form feeds - found 0 form feeds but 7 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** There are 51 instances of lines with control characters in the document. 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. -- Couldn't find a document date in the document -- date freshness check skipped. 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. '1' -- Possible downref: Non-RFC (?) normative reference: ref. '2' -- Possible downref: Non-RFC (?) normative reference: ref. '3' -- Possible downref: Non-RFC (?) normative reference: ref. '4' -- Possible downref: Non-RFC (?) normative reference: ref. '5' Summary: 1 error (**), 0 flaws (~~), 4 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 T2TRG Hong, Choong Seon 2 Internet-Draft Kyung Hee University 3 Intended status: Standards Track Chit Wutyee Zaw 4 Expires: August 09, 2020 Kyung Hee University 5 Kang, Seok Won 6 Kyung Hee University 7 October 2020 9 User Centric Assignment and Partial Task Offloading for Mobile Edge 10 Computing in Ultra-Dense Networks 11 draft-hongcs-chit-mec-00 13 Abstract 15 By collocating servers at base stations, Mobile Edge Computing (MEC) 16 provides low latency to users for real time applications such as 17 Virtual Reality and Augmented Reality. To satisfy the growing demand 18 of users, base stations are deployed densely in highly populated 19 areas. Coordinated Multipoint Transmission (CoMP) allows users to 20 connect to multiple base stations simultaneously. In ultra-dense 21 networks, by offloading the partials of tasks to different base 22 stations, users can achieve lower latency and utilize the computation 23 ability of the surrounding base stations. To control the signaling 24 overhead, the number of base stations that can be connected should be 25 limited. In this paper, we propose a user-centric base station 26 assignment algorithm by considering the possible load of base 27 stations. Moreover, a partial task offloading algorithm is proposed 28 to utilize the computation of under-loaded base stations. Resource 29 allocation is then solved by convex optimization. 31 Status of this Memo 33 This Internet-Draft is submitted in full conformance with the 34 provisions of BCP 78 and BCP 79. 36 Internet-Drafts are working documents of the Internet Engineering 37 Task Force (IETF). Note that other groups may also distribute 38 working documents as Internet-Drafts. The list of current Internet- 39 Drafts is at http://datatracker.ietf.org/drafts/current/. 41 Internet-Drafts are draft documents valid for a maximum of six 42 months and may be updated, replaced, or obsoleted by other 43 documents at any time. It is inappropriate to use Internet-Drafts 44 as reference material or to cite them other than as 45 "work in progress." 46 This Internet-Draft will expire on August 09, 2020. 48 Copyright Notice 50 Copyright (c) 2018 IETF Trust and the persons identified as the 51 document authors. All rights reserved. 53 This document is subject to BCP 78 and the IETF Trust's Legal 54 Provisions Relating to IETF Documents 55 (http://trustee.ietf.org/license-info) in effect on the date of 56 publication of this document. Please review these documents 57 carefully, as they describe your rights and restrictions with respect 58 to this document. Code Components extracted from this document must 59 include Simplified BSD License text as described in Section 4.e of 60 the Trust Legal Provisions and are provided without warranty as 61 described in the Simplified BSD License. 63 Table of Contents 65 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 1 66 1.1. Terminology and Requirements Language . . . . . . . . . . 2 67 2. System Model . . . . . . . . . . . . . . . . . . . . . . . . . . 2 68 3. Problem Formulation. . . . . . . . . . . . . . . . . . . . . . . 4 69 4. User-centric Assignment and Partial Offloading . . . . . . . . . 3 70 4.1. User-centric Assignment. . . . . . . . . . . . . . . . . 3 71 4.2. Partial Offloading . . . . . . . . . . . . . . . . . . . . . . 4 72 4.3. Radio Resource Allocation. . . . . . . . . . . . . . . . . . . 5 73 5. Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 74 6. IANA Considerations. . . . . . . . . . . . . . . . . . . . . . . 6 75 7. Security Considerations . . . . . . . . . . . . . . . . . . . . 6 76 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 77 8.1. Normative References . . . . . . . . . . . . . . . . . . . . . 6 78 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . . 7 80 1. Introduction 82 Mobile Edge Computing (MEC) has been an interesting topic in both 83 academia and industry for its ability to provide low latency and high 84 computation to users by setting up severs near to users. Computation 85 and latency intensive applications requires users to offload their tasks 86 to servers to achieve the minimum delay and maintain the energy of 87 users’ devices. In densely deployed networks, users can utilize the 88 resources of nearby base stations (BS) by offloading partials of their 89 tasks with the technology provided by Coordinated Multipoint 90 Transmission (CoMP). 91 Despite the advantages that MEC brings, there are many challenges to 92 tackle in MEC which are pointed out in [1]. The communication aspect is 93 surveyed in [2] where authors considered joint management of radio and 94 computation resources. Authors also introduced standards and application 95 scenarios. 97 Authors in [3] developed a distributed approach for the offloading of 98 computation tasks, caching of content and allocation of resources by 99 using an alternating direction method of multipliers. Task offloading 100 for ultra-dense network was considered in [4] where authors divided the 101 task placement and resource allocation problems and proposed an 102 efficient offloading approach. But, authors considered to offload to one 103 BS. In this paper, we consider partial offloading in ultra-dense 104 networks. To avoid the overloading at BSs, we take the number of 105 possible users who can connect to BSs into account and propose a 106 heuristic algorithm for user-centric assignment. In addition, a partial 107 offloading algorithm is proposed to utilize the resources of under- 108 loaded BSs by offloading the larger portion of tasks to those BSs. Then, 109 resource allocation is solved with the help of convex optimization. 111 1.1. Terminology and Requirements Language 113 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 114 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 115 document are to be interpreted as described in RFC 2119 [RFC2119]. 117 2. System Model 119 A network with densely deployed BSs is considered where users can 120 offload their tasks to multiple BSs simultaneously. 121 We consider the Orthogonal Frequency Division Multiple Access in both 122 uplink and downlink transmission. We also consider that MEC server are 123 equipped with multi-core technology that they can compute offloaded 124 tasks simultaneously. The user’s task has three parameters such as the 125 size of input file, output result and task in CPU cycles. 127 +---------------+ +--------+ +--------+ +--------+ 128 | Mobile Device | | SBS 1 | | SBS 2 | | SBS 3 | 129 +---------------+ +--------+ +--------+ +--------+ 130 | | 131 | +-----------------+ | 132 | | Offload partial | | 133 | | portion of task | | 134 | +-----------------+ | 135 | +-----------------+ 136 | | Compute the | 137 | | offloaded task | 138 | +-----------------+ 139 | +-----------------+ | 140 | | Return task | | 141 | | result | | 142 | +-----------------+ | 143 --------------------------------------------------------------- 144 | +-----------------+ | 145 | | Offload partial | | 146 | | portion of task | | 147 | +-----------------+ | 148 | +-----------------+ 149 | | Compute the | 150 | | offloaded task | 151 | +-----------------+ 152 | +-----------------+ | 153 | | Return task | | 154 | | result | | 155 | +-----------------+ | 156 --------------------------------------------------------------- 157 | +-----------------+ | 158 | | Offload partial | | 159 | | portion of task | | 160 | +-----------------+ | 161 | +-----------------+ 162 | | Compute the | 163 | | offloaded task | 164 | +-----------------+ 165 | +-----------------+ | 166 | | Return task | | 167 | | result | | 168 | +-----------------+ | 169 --------------------------------------------------------------- 170 Figure 1: Partial offloading with Coordinated Transmission in 171 an Ultra-Dense Network 172 3. Problem Formulation 174 The objective of the partial offloading and resource allocation problem 175 is to minimize the latency of all mobile users where the task must be 176 computed fully. The maximum number of SBSs that a user can associate to 177 is limited. The uplink bandwidth for task offloading and downlink 178 bandwidth for result transmission are limited. In addition, the 179 computing resource at MEC servers and local computing resource are also 180 restricted. 182 4. User-centric Assignment and Partial Offloading 184 4.1. User-centric Assignment to SBSs 185 First, we need to determine the user assignment to the BSs by 186 considering the overloading possibility. The score from a user to a SBS 187 is calculated in which the uplink, downlink singal-to-noise ratios and 188 the inverse proportion of the number of users who are likely to 189 associate to a SBS is considered. 191 +---------------+ +--------+ +--------+ +--------+ 192 | Mobile Device | | SBS 1 | | SBS 2 | | SBS 3 | 193 +---------------+ +--------+ +--------+ +--------+ 194 | 195 +-----------------+ 196 | Calculate score | 197 | for all SBSs | 198 +-----------------+ 199 | 200 +-----------------+ 201 | Choose 3 SBSs | 202 | with highest | 203 | scores | 204 +-----------------+ 205 | 206 --------------------------------------------------------------- 207 | +-----------------+ | 208 | | Send the signal | | 209 | | for assignment | | 210 | +-----------------+ | 211 --------------------------------------------------------------- 212 | +-----------------+ | 213 | | Send the signal | | 214 | | for assignment | | 215 | +-----------------+ | 216 --------------------------------------------------------------- 217 | +-----------------+ | 218 | | Send the signal | | 219 | | for assignment | | 220 | +-----------------+ | 221 --------------------------------------------------------------- 222 Figure 2: User-centric Assignment 224 4.2 Partial Task Offloading 226 After the assignment is done, the fractions of the task allocated to 227 BSs are resolved by utilizing the resources of under-loaded BSs. The 228 higher portion of a task is offloaded to a SBS with a lower total 229 computing load of all the assigned users. SBSs are sorted according 230 to the increasing computing loads of the users. The portion of the 231 task is offloaded to SBSs in the order. 233 +------------------+---------------+------------+------------+ 234 User's| Portion | Portion | Portion | Portion | 235 task | offloaded to | offloaded to |offloaded to| computed at| 236 | SBS 1 | SBS 2 | SBS 3 | the user | 237 +------------------+---------------+------------+------------+ 238 \ / \ / \ / \ / 239 \ / \ / \ / \ / 240 \ / \ / \ / \ / 241 \ / \ / \ / \ / 242 \ / \ / \ / \ / 243 \ / \ / \/ \/ 244 +------------+ +-----------+ +-----------+ +-----------+ 245 | SBS 1 | | SBS 2 | | SBS 3 | | Local | 246 +------------+ +-----------+ +-----------+ +-----------+ 248 Figure 3: Partial Task Offloading 250 4.3. Radio Resource Allocation 252 After obtaining the partial task offloading, we need to solve the 253 resource allocation problem. The resource allocation problem is convex 254 which can easily be solved. In this paper, we use cvxpy [5] to solve 255 this problem. For the local CPU cycles assignment, the maximum 256 available CPU cycle is assigned since the objective is minimizing the 257 latency. 259 5. Results 261 Poisson Point Process is used to model the deployment of BSs and users 262 where their densities are 0.6/m2 6/m2 respectively. For power density 263 thermal noise, -174dBm/Hz is used. Fig. 2 shows the simulation setup 264 used in the paper. Transmit power of pico BSs and users are 23dbm and 265 20dbm respectively. CPU speed is 4GHz at MEC server and 0.3GHz at user. 266 The total uplink and downlink bandwidth are 20MHz each. The size of 267 input file follows a uniform distribution between 300 and 800 KB. The 268 uniform distribution is also used to model the size of tasks and output 269 files which are 0.5 to 1 GHz and 0.2 to 2.5 MB respectively. 270 The latency obtained at SBSs are different but most of the SBSs have the 271 similar latency results due to the different user task requirements. 272 In the highly dense networks, the proposed approach can keep most of the 273 BSs to achieve comparable results. The proposed approach obtains lower 274 latency compared to the baseline approach where the loads of SBSs are 275 not considered and task allocation is done uniformly. The difference 276 becomes significant as the number of users increases. 278 6. IANA Considerations 280 There are no IANA considerations related to this document. 282 7. Security Considerations 284 There are no security considerations related to this document. 286 8. References 287 8.1. Normative References 289 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 290 Requirement Levels", BCP 14, RFC 2119, March 1997. 292 [1] P. Mach and Z. Becvar, "Mobile Edge Computing: A Survey on 293 Architecture and Computation Offloading," IEEE Communications 294 Surveys & Tutorials, vol. 19, no. 3, pp. 1628-1656, 2017. 295 [2] Y. Mao, C. You, J. Zhang, K. Huang and K. B. Letaief, "A 296 Survey on Mobile Edge Computing: The Communication Perspective," 297 IEEE Communications Surveys & Tutorials, vol. 19, no. 4, pp. 298 2322-2358, 2017. 299 [3] C. Wang, C. Liang, F. R. Yu, Q. Chen and L. Tang, "Computation 300 Offloading and Resource Allocation in Wireless Cellular Networks 301 With Mobile Edge Computing," IEEE Transactions on Wireless 302 Communications, vol. 16, no. 8, pp. 4924-4938, 2017. 303 [4] M. Chen and Y. Hao, "Task Offloading for Mobile Edge Computing 304 in Software Defined Ultra-Dense Network," IEEE Journal on 305 Selected Areas in Communications, vol. 36, no. 3, pp. 587-597, 306 2018. 307 [5] S. Diamond and S. Boyd, "CVXPY: A Python-Embedded Modeling 308 Language for Convex Optimization," Journal of Machine Learning 309 Research, vol. 17, no. 83, pp. 1-5, 2016. 311 Authors' Addresses 313 Choong Seon Hong 314 Computer Science and Engineering Department, Kyung Hee University 315 Yongin, South Korea 316 Phone: +82 (0)31 201 2532 317 Email: cshong@khu.ac.kr 319 Chit Wutyee Zaw 320 Computer Science and Engineering Department, Kyung Hee University 321 Yongin, South Korea 322 Phone: +82 (0)31 201 2987 323 Email: cwyzaw@khu.ac.kr 325 Seok Won Kang 326 Computer Science and Engineering Department, Kyung Hee University 327 Yongin, South Korea 328 Phone: +82 (0)31 201 2987 329 Email: dudtntdud@khu.ac.kr