Xu , Qiang , “ Automatic Construction of Coordinated Performance Skeletons ”

The performance skeleton of a parallel application is a short running program whose performance in any scenario reflects the performance of the application it represents. Specifically, the execution time of the performance skeleton is a small fixed fraction of the execution time of the corresponding application in any execution environment. This work presents a framework for automatic construction of performance skeletons. The approach is based on capturing the execution behavior of a parallel application from traces and automatically generating a synthetic skeleton program that reflects that execution behavior. This research introduces coordinated performance skeletons that consist of a single SPMD program that executes on each node of the system.