Novel Approach For Estimating Workload Completion Time Based on Experiment-Driven Modeling and Simulation of the Impact of Inter-Query Interactions

Improving Estimation of Workload Completion Time with Inter-Query Interactions

Abstract

Whileplanning the execution of report-generation workloads, database administratorsoften need to know how long different query workloads will take to run.Database systems run mixes of multiple queries of different types concurrently.Hence, estimating the completion time of a query workload requires reasoningabout query mixes and inter-query interactions in the mixes; rather thanconsidering queries or query types in isolation. This paper presents a novelapproach for estimating workload completion time based on experiment-driven modelingand simulation of the impact of inter-query interactions. A preliminaryevaluation of this approach with TPC-H queries on IBM DB2 shows how ourapproach can consistently predict workload completion times with good accuracy.