Detection and Analysis of Iterative Behavior in Parallel Applications
Many applications exhibit iterative and phase based behavior. We present an approach to detect and analyze iteration phases in applications by recording the control flow graph of the application and analyzing it for loops that represent iterations. Phases are then manually marked and performance profiles are captured in alignment with the iterations. By analyzing how profiles change between capture points, differences in execution behavior between iterations can be uncovered.
KeywordsPhase detection control flow graph continuous profiling
- 1.Fürlinger, K., Gerndt, M.: ompP: A profiling tool for OpenMP. In: Proceedings of the First International Workshop on OpenMP (IWOMP 2005), Eugene, Oregon, USA (2005)Google Scholar
- 4.Casas-Guix, M., Badia, R.M., Labarta, J.: Automatic phase detection of MPI applications. In: Proceedings of the 14th Conference on Parallel Computing (ParCo 2007), Aachen and Juelich, Germany (2007)Google Scholar