The Workload Adaptation in Autonomic DBMSs Based on Layered Queuing Network Model

Abstract

Modern network computing is in the process of changingfrom manual management to autonomic computing. As thecore of network computing, database management systemwith autonomic computing must be able to participate in theprocess to meet the performance requirements of thenetwork applications. This paper studies on workloadadaptation technology. In the Workload AdaptationFramework, the performance forecasting component is thebase of developing workload control scheme. Layeredqueuing network modeling techniques was used to establishthe performance model of DBMS. A test platform based onthis architecture has been built and been tested. Test resultshows that this framework has played an effective role inworkload control, and improved client satisfaction rate andthe application performance of the network.