信息网络安全 ›› 2017, Vol. 17 ›› Issue (7): 11-17.doi: 10.3969/j.issn.1671-1122.2017.07.002

• • 上一篇    下一篇

基于YARN规范的智能电网大数据异常检测

陈阳1(), 王勇2, 孙伟3   

  1. 1. 安徽工程大学现代教育技术中心,安徽芜湖 241000
    2. 安徽工程大学计算机与信息学院,安徽芜湖 241000
    3. 电子科技大学软件学院,四川成都 610054
  • 收稿日期:2017-05-12 出版日期:2017-07-20 发布日期:2020-05-12
  • 作者简介:

    作者简介: 陈阳(1977—),男,安徽,高级工程师,硕士,主要研究方向为大数据应用;王勇(1979—),男,安徽,副教授,博士,主要研究方向为软件工程与机器学习;孙伟(1992—),男,四川,硕士研究生,主要研究方向为密码学。

  • 基金资助:
    国家自然科学基金[61572033];安徽省高校省级自然科学研究重大项目[KJ2015ZD08];安徽省高校优秀人才支持计划重点项目[gxyqZD2016124]

A YARN-based Smart Grid Big Data Abnormal Detection

Yang CHEN1(), Yong WANG2, Wei SUN3   

  1. 1. Center of Modern Education Technology, Anhui Polytechnic University, Wuhu Anhui 241000, China
    2. School of Computer and Information, Anhui Polytechnic University, Wuhu Anhui 241000, China
    3. School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu Sichuan 610054, China
  • Received:2017-05-12 Online:2017-07-20 Published:2020-05-12

摘要:

文章描述了Hadoop早期版本在处理智能电网大数据上的不足,同时分析了YARN规范对Map-Reduce进行改进后对电网大数据处理的优点。文章详细讨论了YARN-DPP平台中智能电网大数据处理的编码与实现及YARN-DPP的优势,并以IEEE 118节点的电网作为智能电网大数据异常检测的案例程序,对YARN-DPP平台的硬件环境与软件运行环境进行配置。实验结果表明,针对海量的智能电网状态安全大数据异常检测的程序,YARN-DPP平台具有较好的吞吐量与加速比,可以满足现代智能电网大数据异常检测的需求,在计算速度上比单机串行计算及Map-Reduce计算要快。

关键词: 智能电网, 大数据, YARN规范, Map-Reduce

Abstract:

The defects of processing smart grid big data in Map-Reduce early version were also discussed, and the advantages of processing smart grid big data in YARN were also described in this paper. The coding model and implementation and advantages of YARN-DPP were also analyzed. In order to demonstrate the effectiveness of YARN-DPP, the hardware configuration environments and software running environments had been completed. A serial of simulation experiments in IEEE 118 node grid system were also done. The results and performance analysis demonstrated that good throughput and speedup had been obtained in YARN-DPP. It can meet the fast demands in large scale grid system big data processing. The computing speed was faster than sequence computation and Map-Reduce computation.

Key words: smart grid, big data, YARN normalization, Map-Reduce

中图分类号: