MABRESE: A New Server Selection Method for Smart SDN-Based CDN Architecture

被引：15

作者：

Hai-Anh Tran ^{[1
]}

Souihi, Sami ^{[2
]}

Duc Tran ^{[1
]}

Mellouk, Abdelhamid ^{[2
]}

机构：

[1] Hanoi Univ Sci & Technol, BKCS Ctr, Hanoi 11615, Vietnam

[2] Univ Paris Est, UPEC, LISSI TincNET EA 3956, F-94400 Vitry Sur Seine, France

来源：

IEEE COMMUNICATIONS LETTERS | 2019年 / 23卷 / 06期

关键词：

SDN; CDN; server selection algorithm; multiarmed bandit;

D O I：

10.1109/LCOMM.2019.2907948

中图分类号：

TN [电子技术、通信技术];

学科分类号：

0809 ;

摘要：

A content delivery network (CDN) is able to handle high traffic and offer reliable services by geographically bringing the content data to the edge network, where replica servers are installed closer to the end users. The traditional CDN architecture lacks a global view of the whole network and, thus, cannot dynamically and optimally perform the server selection task. In this letter, we propose to integrate a software-defined network (SDN) into the CDN architecture. Our intuitive motivation is to decouple the control plane from the forwarding plane in order to offer flexibility and programmability by using the centralized controller. This letter also develops a novel server selection algorithm that is based on the so-called multi-armed bandit problem. Such an algorithm is proved to optimize the server selection function and provide a good experimental result in terms of both average response time and reward score.

引用

页码：1012 / 1015

页数：4

共 10 条

[1]

[Anonymous], 2017, IEEE INT C COMM ICC, DOI DOI 10.1109/ICC.2017.7996499

[2]

[Anonymous], ONOS SECURITY PERFOR

[3] On the Impact of Network State Collection on the Performance of SDN Applications [J].