Cloud-Based Concurrent Simulation at Work: Fast Performance Prediction of Parallel Programs

The performance prediction of parallel programs is a time-consuming process due to the need to execute a large number of simulations for different values of parameters linked to the software decomposition and to the target hardware environment. This paper shows how mJADES, a novel environment for running concurrent simulations in the cloud, can provide an effective support to a simulation-based performance prediction framework. The accuracy of predictions and the practical validity of the approach followed are evaluated through experimentation on a simple but realistic case study.