A Methodology for Bridging the Native and Simulated Execution of Parallel Applications

Simulation is considered as the third pillar of science, following experimentation and theory. Bridging the native and simulated executions of parallel applications is needed for attaining trustworthiness in simulation results. Yet, bridging the native and simulated executions of parallel applications is challenging. This work proposes a methodology for bridging the native and simulated executions of message passing parallel applications on high performance computing (HPC) systems in two steps: Expression of the software characteristics, and representation and verification of the hardware characteristics in the simulation. This work exploits the capabilities of the SimGrid [3] simulation toolkit's interfaces to reduce the effort of bridging the native and simulated executions of a parallel application on an HPC system. For an application from computer vision, the simulation of its parallel execution using straightforward parallelization on an HPC cluster approaches the native performance with a minimum relative percentage difference of 5.6%.