电子学报 ›› 2019, Vol. 47 ›› Issue (10): 2061-2068.DOI: 10.3969/j.issn.0372-2112.2019.10.006

• 学术论文 • 上一篇    下一篇

基于松弛时间与累计发送量的数据中心网络混合流调度机制

臧韦菲, 兰巨龙, 胡宇翔   

  1. 国家数字交换系统工程技术研究中心, 河南郑州 450002
  • 收稿日期:2017-12-27 修回日期:2019-02-28 出版日期:2019-10-25
    • 作者简介:
    • 臧韦菲 女,1993年出生,河南郑州人.国家数字交换系统工程技术研究中心博士研究生.主要研究方向为新型网络体系结构.E-mail:zwfndsc@163.com;兰巨龙 男,1962年出生,河北张北人.国家数据交换系统工程技术研究中心教授、博士生导师.主要从事新一代信息网络关键理论与技术的研究工作.
    • 基金资助:
    • 国家网络空间安全专项课题 (No.2017YFB0803204); 国家863高技术研究发展计划 (No.2015AA016102); 国家自然科学基金创新研究群体科学基金 (No.61521003)

Slack Time and Accumulation-Based Mix-flow Scheduling in Data Center Networks

ZANG Wei-fei, LAN Ju-long, HU Yu-xiang   

  1. National Digital Switching System Engineering & Technological Research Center, Zhengzhou, Henan 450002, China
  • Received:2017-12-27 Revised:2019-02-28 Online:2019-10-25 Published:2019-10-25
    • Supported by:
    • National Cyber Space Security Project of China (No.2017YFB0803204); National High-tech R&D Program of China  (863 Program) (No.2015AA016102); NSFC Innovation Research Group (No.61521003)

摘要: 数据中心网络中同时存在截止时间流(deadline flow)和非截止时间流(non-deadline flow),为降低非截止时间流的平均完成时间(Average Flow Complete Time,AFCT)同时维持低截止时间错失率(Deadline Miss Rate,DMR),本文提出了一种基于松弛时间与累计发送量的混合流调度机制(Slack Time and Accumulation based Mix-flow Scheduling,STAM).首先通过引入松弛时间的概念,衡量截止时间流对非截止时间流在传输时延上的宽容度;然后根据松弛时间,通过使截止时间流尽可能接近其规定截止时间完成,降低非截止时间流的完成时间;最后,利用最小累计发送量优先策略进一步降低非截止时间流的平均完成时间.仿真结果表明,该机制能有效降低非截止时间流的平均完成时间,同时保证较低的截止时间错失率.

关键词: 数据中心网络, 流量调度, 截止时间, 松弛时间

Abstract: Applications deployed in data center networks generate a mix of flows with and without deadlines.To reduce the average flow complete time (AFCT) while maintain a low deadline miss rate (DMR),a slack time and accumulation-based mix-flow scheduling mechanism (STAM) is proposed in this paper.Firstly,the delay tolerance of deadline flows on non-deadline flows is measured with the introduction of Slack Time.Then the cost of complete time of non-deadline flows is reduced by completing deadline flows barely before their deadlines.Lastly,non-deadline flow is scheduled according to the number of bytes it has sent to reduce the average flow completion time.Experiment results show that the proposed mechanism can effectively reduce the average flow completion time of non-deadline flows while maintaining low deadline miss rate.

Key words: data center network, flow scheduling, deadline, slack time

中图分类号: