idnits 2.17.1 draft-chong-t2trg-llwnc-00.txt: -(214): Line appears to be too long, but this could be caused by non-ascii characters in UTF-8 encoding 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 8 instances of lines with non-ascii characters in the document. == The page length should not exceed 58 lines per page, but there was 1 longer page, the longest (page 2) being 59 lines Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** There is 1 instance of too long lines in the document, the longest one being 1 character in excess of 72. ** There are 5 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 date (October 2020) is 1290 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) No issues found here. Summary: 2 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 T2TRG Chong, Song 2 Internet-Draft KAIST 3 Intended status: Standards Track Jang, Hyeonjoon 4 Expires: December 15, 2021 KAIST 5 October 2020 7 Low End-to-End Latency Content Caching 8 in Wireless Network Clouds 9 draft-chong-t2trg-llwnc-00 11 Abstract 13 In this document, we consider the content caching design without 14 requiring historical content access information or content popularity 15 profiles in a hierarchical cellular network architecture. 16 Our design aims to dynamically select caching locations for 17 different contents where caching locations can be content servers, 18 cloud units (CUs), and base stations (BSs). Our design objective 19 is to support as high content request rates as possible while 20 maintaining the finite service time. 22 Status of this Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at http://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six 33 months and may be updated, replaced, or obsoleted by other 34 documents at any time. It is inappropriate to use Internet-Drafts 35 as reference material or to cite them other than as 36 "work in progress." 38 This Internet-Draft will expire on December 15, 2021. 40 Copyright Notice 42 Copyright (c) 2020 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (http://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . .. . . . . . 2 58 2. Main Idea . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 2.1. System Model . . . . . . . . . . . . . . . . . . . . . . . . . 3 60 2.2. Hybrid Content Caching Design . . . . . . . . . . . .. . . . . 4 61 3. IANA Considerations . . . . . . .. . . . . . . . . . . . . . . 5 62 4. Security Considerations . . . . . . . . . . . . . . . . . . . 5 63 5. References . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 5.1. Normative References . . . . . . . . . . . . . . . . . . . . . 5 65 5.2. Informative References . . . . . .. . . . . . . . . . . . . . 5 66 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . 5 67 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . . 6 69 1. Introduction 71 With the rapidly increasing mobile video traffic, 72 both backhaul and fronthaul networks connecting the internet 73 with the mobile core and edge networks such as base stations 74 (BSs) (see Fig. 1) become more and more congested. Since 75 popular contents are more frequently requested by end users, 76 we can reduce the end-to-end latency of video content services 77 and backhaul/fronthaul traffic loads by placing popular 78 contents at locations closer to the end users. Content 79 popularity profile, which captures its average access frequency 80 from end users, can be spatio-temporally varying in wireless 81 networks due to user mobility or social interaction between 82 mobile users. Moreover, the massive deployment of small 83 cells such as femto-cells in the cellular networks increases 84 the spatial granularity which renders the real-time estimation 85 of the content popularity in each BS more challenging. 86 Meanwhile, edge-centric technologies (e.g., edge computing/ 87 caching and distributed Self-Organizing Network (SON)) 88 and cloud-centric technologies (e.g., cloud computing/caching 89 and Cloud Radio Access Network (C-RAN)) have been devised 90 to support the latency-critical applications and to address 91 the huge amount of workloads in the cloud servers, respectively. 92 Recently, hybrid network architecture and operations 93 which adaptively exploit the edge-centric and cloud-centric 94 natures of the wireless network environments, has been proposed. 96 2. Main Idea 98 In this document we consider the caching design in a general wireless 99 network architecture where each group of a small number of BSs is 100 connected to a cloud unit(CU) where individual BSs and the CU have 101 content caching repositories (see Fig. 1). The caching design must 102 adaptively determine the content caching locations at BSs and CUs 103 depending on the network dynamics to support as high average content 104 request rates as possible with finite content service time. To address 105 this design while not requiring information on historical content access 106 or content popularity profile, we employ the Lyapunov optimization 107 method [a] for which the short-term max-weight problem derived from the 108 Lyapunov drift must be optimized in the slot-by-slot basis. Because the 109 max-weight problem is NP-hard and difficult to tackle [b] due to the 110 coupling between CU and BS caching decisions, we propose an 111 approximation algorithm which achieves the constant approximation ratio 112 to the optimal weight by exploiting the submodularity of the 113 slot-by-slot objective function and the structure of hierarchical 114 content caching networks. 116 +------------------+ +-------+ +--+ End-to-End +------+ 117 | Original Content |---------| Cloud |--------|BS| Path(case1) | End | 118 | Servers | Backhaul| Unit | Frount +--+ <==========>| | 119 +------------------+ +-------+ haul | users| 120 | <=================================================>+------+ 121 Backhaul | End-to-End Path(case3) 122 | 123 +-------+ Frouthaul +--+ +------+ 124 | Cloud |----------- |BS| ------------| End | 125 | | +--+ | users| 126 | Unit |<===========================>+------+ 127 +-------+ End-to-End Path(case2) 129 Figure 1: Network Architecture 131 2.1 System Model 133 Fig. 1 illustrates the hierarchical network architecture considered in 134 the dynamic caching design. We consider a video content set where 135 the file size of different contents is assumed to be the same and equal 136 to s (in bits). Each video content file is assumed to be split into 137 multiple chunks, and we assume that a file can be recovered at 138 the destination even if not all of the chunks for the file are 139 successfully delivered [c]. We assume there are E cloud units (CUs) and 140 N base stations (BSs) in the system. All contents are saved in their 141 own original content servers distributed throughout the internet. 142 We consider a time-slotted system with equal-size time slots of 143 duration Δt and the time slots are indexed as t = 0, 1, ...2. 144 Moreover, caching control decisions are made in the slot-by slot basis. 145 In each time slot t, the amount of data from content requested by end 146 users associated with BSs and CUs is independent and identically 147 distributed over time slots. 149 When content f is requested from an end user, the requested content is 150 transmitted from its closest caching location among the associated BS, 151 CU and original content server which has the corresponding content 152 placed by a deployed caching strategy. Hereafter, we call this closest 153 network node as a source node of content f. Hence, the end-to-end path 154 of content f could vary as the choice of the source node changes. 155 (Figure 1.) 157 To capture the dynamics of content requests, services, and the 158 corresponding content service time, we introduce virtual queues for 159 each BS where the virtual queue backlog for content f, 160 i.e., Q_f(t), evolves over time as follows: 162 Q_f(t+1) = [Q_f(t) - r_f(t) + A_f(t)]^+ , 164 for every CU and associated BS, content(f), where [x]^+ = max(x,0). 165 In the above, we define A_f(t) and r_f(t) as the amount of data 166 from content f requested by end users associated with BSs and the 167 amount of served data of content f at BSs, respectively. 168 The amount of served data r_f(t) from each virtual queue during time slot 169 t depends on the caching decision, i.e., to cache content f at CU or BS. 170 The average virtual queue backlog of content f indirectly captures the 171 average end-to-end latency of content f. 173 2.2. Hybrid Content Caching Design 175 The hybrid content caching algorithm should (i) achieve low average 176 end-to-end latency by stabilizing virtual queues if the content request 177 rate vector is within the capacity region and (ii) support as high 178 average content request rates as possible. For any content request rate 179 vector inside the capacity region,all virtual queues must be stable, 180 i.e., supremum of the following value 182 t-1 183 (1/r)* Σ {expectation of sum of Q_f(r) w.r.t every f, BS and CU} 184 r=0 186 should be bounded as t goes to infinity. 187 To develop such a caching algorithm, we could employ the Lyapunov 188 optimization method [a]. Toward this end, we define Lyapunov function 189 and Lyapunov drift function w.r.t. Q_f(t) and derive an upper bound of 190 the Lyapunov drift function using the queueing dynamics of the virtual 191 queues. Then the content caching algorithm can be developed by 192 minimizing the upper bound of the Lyapunov drift function in each time 193 slot. 195 3. IANA Considerations 197 There are no IANA considerations related to this document. 199 4. Security Considerations 201 There are no security considerations related to this document. 203 5. References 205 5.1. Normative References 207 [a] M. Neely, “Stochastic network optimization with application to 208 communication and queueing systems,” Synthesis Lectures on 209 Communication Networks, pp. 1–211, 2010. 211 [b] R. Hemmecke, M. Koppe, J. Lee, and R. Weismantel, 50 Years of 212 Integer Programming 1958-2008. Springer, 2009. 214 [c] F. Pantisano, M. Bennis, W. Saad, and M. Debbah, “Match to Cache 215 :Joint user association and backhaul allocation in cache-aware 216 small cell networks,” in Proc. of IEEE ICC, London, UK, Jun. 217 2015, pp. 3082–3087. 219 5.2. Informative References 221 Acknowledgements 223 This work was supported by Institute for Information & communications 224 Technology Promotion(IITP) grant funded by the Korea government(MSIT) 225 (No.2015-0-00557, Resilient/Fault-Tolerant Autonomic Networking Based 226 on Physicality, Relationship and Service Semantic of IoT Devices) 227 Authors' Addresses 229 Song Chong 230 The Graduate School of Artificial Intelligence, 231 Korea Advanced Institute of Science and Technology(KAIST) 232 Daejeon, South Korea 233 Phone: +82 (0)42 350 3473 234 Email: songchong@kaist.edu 236 Hyeonjoon Jang 237 Electrical Engineering Department, 238 Korea Advanced Institute of Science and Technology(KAIST) 239 Daejeon, South Korea 240 Phone: +82 (0)42 350 5473 241 Email: thefelix@kaist.ac.kr 243 Sewoong Lee 244 Electrical Engineering Department, 245 Korea Advanced Institute of Science and Technology(KAIST) 246 Daejeon, South Korea 247 Phone: +82 (0)42 350 5473 248 Email: dltpdnd21@kaist.ac.kr