Determining suitable execution environment based on dynamic workload during early stages of software development life cycle: a simulation approach

Abstract

In the early development stages of distributed software systems, assessing performance and corresponding prediction techniques remains an important but largely unsolved problem. In the distributed environment, the workload and the deployment environment of the software components have major impact on the performance and they are uncertain. The problem of uncertainty regarding performance requirement data is required to be addressed by estimation techniques. Hence, in this paper we present a mathematical model for performance calculation and simulate the model that supports a decision maker involved in performance analysis for a given distributed application. We analyse the execution environment by considering various configurations of hardware resources based on dynamic workload over a time horizon. The methodology is demonstrated through a case study and the results are validated. Results show that the proposed methodology is useful in identifying suitable execution environment corresponding to the defined performance goal. The bottleneck resources are identified and suggestions for improvement are given.