计算机技术与发展Issue(1)：70-73,4.DOI:10.3969/j.issn.1673-629X.2016.01.014
MapReduce 性能预测模型构建
Performance Prediction Model Construction of MapReduce
摘要
Abstract
MapReduce is the most popular cloud computing model in big data processing. Predicting the performance of MapReduce could be used to increase the cloud computing efficiency. However,MapReduce runs based on a huge number of configuration parameters which would affect the performance. Traditional predicting of configuration is based on the experience of administrator,and this approach is of low accuracy. In order to give a better prediction of MapReduce performance,a multiple linear regression model based on the configura-tion parameters was proposed. With the aim to verify the model,an experiment was carried out taking the Map number and Reduce num-ber as an example. The experiments results indicate that the proposed model can be used in predicting the MapReduce performance.关键词
MapReduce/云计算模型/性能预测/多元线性回归模型Key words
MapReduce/cloud computing model/performance prediction/multiple linear regression model分类
信息技术与安全科学引用本文复制引用
李振举,李学军,刘涛,杨晟..MapReduce 性能预测模型构建[J].计算机技术与发展,2016,(1):70-73,4.基金项目
总装备部预研项目(513150701) （513150701）
