节点文献
基于海量数据的Web集群的性能分析与测试
Performance Analysis and Testing of Web Cluster Base on Mass Data
【摘要】 为解决海量数据和Web集群服务的下载配置模型测试问题,通过测试工具选择、测试环境配置、测试用例选择以及测试实施过程,利用性能测试方法,针对一组真实测试实例,得出与现有的单机普通阵列模型相比,采用新的集群模型后,一台高性能的磁盘阵列连接4台服务器,1 300个用户并发下载时,阵列CPU使用率持平在90%的情况下,平均每秒能完成360个文件的下载任务,达到了阵列的最大服务能力,为目前最佳集群下载配置模型。
【Abstract】 The problem of testing of download config model will be explored.The method of performance testing comprise the selection of testing tools,configuration of testing environment,selection of test case and progress of test execution.A right cluster downloading config model will be concluded,consists of a higher performance disk array could connect four servers,the usage of array CPU keep about 90%,meanwhile download is implementing by 1 300 users simultaneously,in such case the task of 360 files download could be accomplished.
【Key words】 cluster server; Web service; mass data; performance analysis; performance testing;
- 【文献出处】 吉林大学学报(信息科学版) ,Journal of Jilin University(Information Science Edition) , 编辑部邮箱 ,2007年06期
- 【分类号】TP393.09
- 【被引频次】8
- 【下载频次】238