Easy-to-use SAP sizing based on evolutionary generated scalability models

abstract

Sizing an SAP ERP system's hardware is a complex task, due to the numerous factors affecting the system?s performance and scalability. To ease the effort required for the sizing process, we developed the SAP Sizing Calculator (SC) tool. The SC requires a minimum set of input data and provides a fast prediction of the estimated system performance. The prediction algorithm is based on the results of a synthetic benchmark, the Zachmanntest, which we applied to various x86 servers. From the numerous Zachmanntest results we extracted scalability models using evolutionary algorithms, which we integrated into the SC. Beside its simple application the SC provides an extension interface for easy integration of scalability models derived from SAP systems deployed on other server architectures.