基于云计算的并行数据挖掘系统设计与实现  

Parallel Data Mining System Design and Implementation Based on Cloud Computing

作　　者：李凯[1] 常征[1] 

机构地区：[1]成都信息工程学院计算机学院,成都610225

出　　处：《微计算机信息》2011年第6期121-123,共3页Control & Automation

摘　　要：为了提供虚拟化、高扩展和高性能的并行数据挖掘平台,提出了基于云计算平台的并行数据挖掘模型。文章中首先在GoogleAppEngine平台的基础上扩展了数据规约功能,然后提出了并行数据挖掘平台的系统构架和层次结构,最后给出了平台中各功能模块的具体设计和实现细节。实验分析表明基于云计算平台的并行数据挖掘系统执行效率要比单机系统高,并随着数据量的增大,效率优势越明显。In order to provide a virtualization,high scalability and high-performance of data mining platform,based on the cloud computing,a data mining platform mode in parallel is proposed in this paper.Firstly,on the basis of？Google App Engine development platform,the function of data reduction is expanded.After that,the system architecture of the data mining platform in parallel？ based on cloud computing is presented.Finally in this paper,the design thought？and implementation technique of each function module is put forward.Based on the experimental analysis,the cloud of parallel computing platform for data mining system efficiency than single system is high,and with the increasing amount of data,the more obvious efficiency advantage.

关 键 词：并行数据挖掘 云计算 虚拟化 GoogleAppEngine 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]