Study of Different Intelligence Level of Workload Within Completion Times |

ABSTRACT:

While planningthe execution of report-generation workloads, database administrators oftenneed to know how long different query workloads will take to run. Databasesystems run mixes of multiple queries of different types concurrently. Hence,estimating the completion time of a query workload requires reasoning aboutquery mixes and inter-query interactions in the mixes; rather than consideringqueries or query types in isolation. This paper presents a novel approach forestimating workload completion time based on experiment-driven modeling andsimulation of the impact of inter-query interactions. A preliminary evaluationof this approach with TPC-H queries on IBM DB2 shows how our approach canconsistently predict workload completion times with good accuracy.