采用Map-Reduce模型的海量电能质量数据交换格式文件快速解析方案.pdfVIP

38 卷 6 期 电 网 技 术 Vol. 38 No. 6 20 14 年6 月 Power System Technology Jun. 2014 文章编号：1000-3673 （20 14）06-1705-07 中图分类号：TM 72 1 文献标志码：A 学科代码：470 4051 采用Map-Reduce 模型的海量电能质量 数据交换格式文件快速解析方案 1 1 2 曲广龙 ，杨洪耕 ，张逸 1．四川大学 电气信息学院，四川省 成都市 610065； 2．国网福建省电力有限公司电力科学研究院，福建省 福州市 350000 ） A Fast Parallel Parsing Scheme for Massive PQDIF Files With Map-Reduce Model 1 1 2 QU Guanglong , YANG Honggeng , ZHANG Yi (1. School of Electrical Engineering and Information, Sichuan University, Chengdu 610065, Sichuan Province, China; 2. Fuji an Electric Power Research Institute, Fuzhou 350000, Fuji an Province, China) ABSTRACT: To solve the problem of low parsing efficiency DOI 10.13335/j.1000-3673.pst.2014.06.045 and long parsing time caused by massive power quality data 0 引言 interchange format (PQDIF) files in the present power quality monitoring system, a fast parsing scheme for massive PQDIF 电能质量数据交换格式(power quality data files based on cluster computing architecture was researched interchange format，PQDIF)[1]是IEEE P 1159.3 标准 and implemented. Fast parallel parsing of the massive PQDIF 中规定的一种通用电能质量数据交换格式，它能完 files was realized in this scheme with Map-Reduce model. To 全独立于监测设备的软、硬件，使数据具有良好兼 improve operation efficiency, Map-Reduce model split the 容性，既可实现不同监测装置或监测系统间数据共 large-scale operation task of data sets into several subtasks and 享，还具备多角度观察电能质量物理属性的特性， assign them to each compute node to process on