Research On Performance Testing And Optimization Based On Massive Data Storage

With the expansion of the scale of data storage, the traditional arrays of disks technology and network storage technology have already could not satisfy the performance requirements of massive data storage, high-performance cloud storage technology is gradually replacing the traditional data storage technology. In order to meet the performance requirements of massive data storage system, this paper uses the Hadoop distributed storage framework which is on behalf of cloud storage service, combining with the non-relational database HBase, and distributed file system HDFS, and parallel programming model MapReduce, to build and deploy Hadoop cluster for the inventory and logistics distribution management system, and to meet the system’s distributed, large capacity, high performance and high scalability requirements. In strict accordance with the process of performance testing, using the automated testing tools LoadRunner to do system performance testing, and making up test results, and analysising the system’s performance bottlenecks, and finally combining with HBase database optimization, Hadoop cluster optimization, Web server upgrade and job scheduling strategy improvement, to achieve the goal of system performance optimization.