Budget-Constrained Edge Service Provisioning With Demand Estimation via Bandit Learning

被引:35
作者
Chen, Lixing [1 ]
Xu, Jie [1 ]
机构
[1] Univ Miami, Dept Elect & Comp Engn, Coral Gables, FL 33146 USA
关键词
Edge computing; service placement; multi-armed bandit; ALGORITHM; CONFLICT;
D O I
10.1109/JSAC.2019.2933781
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Shared edge computing platforms, which enable Application Service Providers (ASPs) to deploy applications in close proximity to mobile users are providing ultra-low latency and location-awareness to a rich portfolio of services. Though ubiquitous edge service provisioning, i.e., deploying the application at all possible edge sites, is always preferable, it is impractical due to often limited operational budget of ASPs. In this case, an ASP has to cautiously decide where to deploy the edge service and how much budget it is willing to use. A central issue here is that the service demand received by each edge site, which is the key factor of deploying benefit, is unknown to ASPs a priori. What's more complicated is that this demand pattern varies temporally and spatially across geographically distributed edge sites. In this paper, we investigate an edge resource rental problem where the ASP learns service demand patterns for individual edge sites while renting computation resource at these sites to host its applications for edge service provisioning. An online algorithm, called Context-aware Online Edge Resource Rental (COERR), is proposed based on the framework of Contextual Combinatorial Multi-armed Bandit (CC-MAB). COERR observes side-information (context) to learn the demand patterns of edge sites and decides rental decisions (including where to rent the computation resource and how much to rent) to maximize ASP's utility given a limited budget. COERR provides a provable performance achieving sublinear regret compared to an Oracle algorithm that knows exactly the expected service demand of edge sites. Experiments are carried out on a real-world dataset and the results show that COERR significantly outperforms other benchmarks.
引用
收藏
页码:2364 / 2376
页数:13
相关论文
共 39 条
[1]  
[Anonymous], 2017, P 2 ACM IEEE S EDG C
[2]   Finite-time analysis of the multiarmed bandit problem [J].
Auer, P ;
Cesa-Bianchi, N ;
Fischer, P .
MACHINE LEARNING, 2002, 47 (2-3) :235-256
[3]  
Bergman N., 1999, THESIS, V579, P11
[4]   A Branch-and-Bound Algorithm for the Knapsack Problem with Conflict Graph [J].
Bettinelli, Andrea ;
Cacchiani, Valentina ;
Malaguti, Enrico .
INFORMS JOURNAL ON COMPUTING, 2017, 29 (03) :457-473
[5]   How to use expert advice [J].
CesaBianchi, N ;
Freund, Y ;
Haussler, D ;
Helmbold, DP ;
Schapire, RE ;
Warmuth, MK .
JOURNAL OF THE ACM, 1997, 44 (03) :427-485
[6]  
Chen L., 2019, ARXIV190309080
[7]   Spatio-Temporal Edge Service Placement: A Bandit Learning Approach [J].
Chen, Lixing ;
Xu, Jie ;
Ren, Shaolei ;
Zhou, Pan .
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2018, 17 (12) :8388-8401
[8]   Computation Peer Offloading for Energy-Constrained Mobile Edge Computing in Small-Cell Networks [J].
Chen, Lixing ;
Zhou, Sheng ;
Xu, Jie .
IEEE-ACM TRANSACTIONS ON NETWORKING, 2018, 26 (04) :1619-1632
[9]   A Dynamic Pooling Approach to Extract Complete Allele Signal Information in Somatic Copy Number Alternations Detection [J].
Cheng, Long ;
Yao, Pengfei ;
Lu, Jianwei ;
Hao, Ke ;
Zhang, Zhongyang .
PROCEEDINGS OF 2018 6TH INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND COMPUTATIONAL BIOLOGY (ICBCB 2018), 2018, :1-6
[10]   A fine OPO solid state laser with heat-conduction cooling element [J].
Cheng, Y ;
Ding, FZ ;
Sun, B ;
Wang, XB ;
Wang, GC ;
Li, Y ;
Hu, C ;
Wu, YG .
SEMICONDUCTOR LASERS AND APPLICATIONS, 2002, 4913 :306-311