计算机应用 ›› 2012, Vol. 32 ›› Issue (03): 595-598.DOI: 10.3724/SP.J.1087.2012.00595

• 先进计算 •    下一篇

基于并行机制的商务智能系统BI-PaaS

江志雄1,2, 金海1, 黄晓庆2   

  1. 1.华中科技大学 计算机科学与技术学院,武汉 430074;

    2.中国移动通信研究院,北京100053

  • 收稿日期:2011-09-07 修回日期:2011-11-21 发布日期:2012-03-01 出版日期:2012-03-01
  • 通讯作者: 江志雄
  • 作者简介:江志雄(1975-),男,江西上饶人,博士,CCF会员,主要研究方向:基于云计算的海量信息分析;金海(1966-),上海人,教授,博士生导师,博士,CCF高级会员,主要研究方向:网格计算、计算系统虚拟化;黄晓庆(1962-),男,重庆人,硕士,主要研究方向:移动通信、手机操作系统、云计算基础设施。
  • 基金资助:

    国家科技重大专项(2008ZX03006);中国移动研究院云计算计划单列项目(879011R102101600)。

BI-PaaS: parallel-based business intelligence system

JIANG Zhi-xiong1,2, JIN Hai1, HUANG Xiao-qing2   

  1. 1.School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan Hubei 430074, China;

    2.China Mobile Research Institute, Beijing 100053, China

  • Received:2011-09-07 Revised:2011-11-21 Online:2012-03-01 Published:2012-03-01

摘要: 针对海量信息处理对传统的商业智能软件提出的挑战,设计并实现了一种基于并行机制的商务智能(BI)系统(BI-PaaS)。该系统搭建在中国移动大云基础设施之上, 以云计算的强大并行计算和分布式存储能力为支撑,将ETL、DM、OLAP、Report等各类BI能力并行化。实验结果表明,基于并行机制的BI-PaaS系统可以大大提高BI能力的数据处理性能,有效地支持电信运营的海量数据分析。

关键词: 并行化, PaaS, 商务智能, 云计算, 数据挖掘, 分布式存储

Abstract: Concerning the challenge of massive information to the traditional Business Intelligence (BI) system, the prototype of a BI system (BI-PaaS) based on parallel mechanism was designed and implemented. The structure was built upon the project named big cloud in China mobile, being powered by massively parallel computing and distributed storage and integrated with several technologies in terms of ETL, DM, OLAP, Report, etc. The results of the experiment prove that the function based upon parallel computing greatly raises the competency of data processing and effectively supports the data analysis.

Key words: parallelization, Platform as a Service (PaaS), Business Intelligence (BI), cloud computing, Data Mining (DM), distributed storage

中图分类号: