Supporting capacity planning for DB2 UDB

Capacity planning is the process of determining the most cost-effective computing environment that meets the current and future demands of a computing system. Capacity planning is important for large database management systems (DBMSs) whose performance can be greatly affected by the amount of available resources and whose workloads can change significantly over time. In this paper, we examine support for capacity planning for DBMSs. We describe an analytical model that is used to estimate performance under different scenarios and discuss how to obtain the input parameters to the model for an OLTP workload running on IBM® DB2® Universal DatabaseTM. We then give an example of how the model can be used to solve a typical capacity planning problem.