计算机科学与探索 ›› 2021, Vol. 15 ›› Issue (2): 261-269.DOI: 10.3778/j.issn.1673-9418.2003024

• 网络与信息安全 • 上一篇    下一篇

数据中心网络中基于ELM的流簇大小推理机制

叶进,谢紫琪,肖庆宇,宋玲,李晓欢   

  1. 1. 广西多媒体通信与网络技术重点实验室(广西大学 计算机与电子信息学院),南宁 530004
    2. 桂林电子科技大学 信息与通信学院,广西 桂林 541004
    3. 综合交通大数据应用技术国家工程实验室(广西),南宁 530001
  • 出版日期:2021-02-01 发布日期:2021-02-01

Inferring Coflow Size Mechanism Based on ELM in Data Center Network

YE Jin, XIE Ziqi, XIAO Qingyu, SONG Ling, LI Xiaohuan   

  1. 1. Guangxi Key Laboratory of Multimedia Communications and Network Technology (School of Computer and Electronic Information, Guangxi University), Nanning 530004, China
    2. School of Information and Communication, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China
    3. National Engineering Laboratory for Comprehensive Transportation Big Data Application Technology (Guangxi), Nanning 530001, China
  • Online:2021-02-01 Published:2021-02-01

摘要:

近年来研究流簇(Coflow)为单位的调度策略成为改进数据中心网络的新热点。然而现有的信息未知流簇调度器难以快速地推理任务级信息,导致小任务不能被及时调度,以及平均任务完成时间无法最小化。因此数据中心网络需要更加高效的推理模型提升流簇大小判断的准确性和敏感性。提出了一种基于机器学习的流簇大小推理模型(MLcoflow),利用极限学习机(ELM)以最小训练误差为求解目标建立推理模型,并且使用不完全信息建模以提升敏感度。实验证明与其他算法相比,ELM方法的准确性评分平均高出19.8%,敏感度平均高出10.2%。通过仿真模拟对比了几种调度器,基于MLcoflow的调度器将平均任务完成时间降低了20.1%。

关键词: 数据中心, 流簇大小, 流簇调度, 推理模型, 极限学习机(ELM)

Abstract:

In recent years, Coflow scheduling has become a research hotspot in data center network. However, it is difficult for existing non-clairvoyant Coflow schedulers to infer the task information quickly. Therefore, small tasks cannot be scheduled in time, making it fail to minimize the average task completion time. Data center network requires effective inferring model to improve the accuracy and sensitivity in inferring Coflow size. This paper proposes a machine learning based Coflow size inferring model (MLcoflow), which utilizes an extreme learning machine (ELM) to establish Coflow size inferring model to minimize training error, and uses the incomplete infor-mation in training to increase the sensitivity. Experiment results show that the accurate score and sensitivity of ELM method are 19.8% and 10.2% higher than other algorithms on average, respectively. This paper compares several schedulers by simulation. MLcoflow-based scheduler reduces the average task completion time by up to 20.1%.

Key words: data center, Coflow size, Coflow scheduling, inferring model, extreme learning machine (ELM)