Scalable Workload Adaptation for Mixed Workload

Workload adaptation is a performance management process in which an autonomic database management system (DBMS) efficiently makes use of its resources by filtering or controlling the workload presented to it in order to meet its Service Level Objectives (SLOs). The overhead incurred by filtering or controlling the workload is an important factor affecting the effectiveness of workload adaptation. This paper investigates the overhead of AWMF, a framework for workload adaptation and proposes a scalable approach for adapting mixed workload under the framework. The proposed approach allows Query Scheduler, the prototype implementation of AWMF, manage both OLAP and OLTP classes of queries to meet their performance goals by allocating DBMS resources through admission control in the presence of workload fluctuation. Experiments with IBM® DB2® Universal DatabaseTM are conducted to show the proposed approach is scalable and effective.