PAS2P Tool, Parallel Application Signature for Performance Prediction
Accurate prediction of parallel applications’ performance is becoming increasingly complex. We seek to characterize the behavior of message-passing applications by extracting a signature to predict the performance in different target systems. We have developed a tool we called Parallel Application Signature for Performance Prediction (PAS2P) that strives to describe an application based on its behavior. Based on the application’s message-passing activity, we have been able to identify and extract representative phases, with which we created a signature. We have experimented using scientific applications and we predicted the execution times on multicore architectures with an average accuracy of over 97%.
KeywordsPerformance Prediction Parallel Application Signature
Unable to display preview. Download preview PDF.
- 1.Bailey, D., Barszcz, E., Barton, J., Browning, D.: The NAS Parallel Benchmarks. International Journal of High Performance Computing (January 1991)Google Scholar
- 2.Brown, P.N., Falgout, R.D., Jones, J.E.: Semicoarsening multigrid on distributed memory machines. SIAM Journal on Scientific Computing 21, 1823–1834 (2000)Google Scholar
- 4.Gustafson, J., Snell, Q.: Hint: A new way to measure computer performance. In: Hawaii International Conference on System Sciences, p. 392 (1995)Google Scholar
- 6.Hoisie, A., Lubeck, O., Wasserman, H.: Performance and scalability analysis of teraflop-scale parallel architectures using multidimensional. Journal of High Performance Computing Applications (January 2000)Google Scholar
- 7.Hursey, J., Squyres, J.M., Lumsdaine, A.: A checkpoint and restart service specification for open mpi. Technical Report TR635, Indiana University, Bloomington, Indiana, USA (July 2006)Google Scholar
- 9.Sherwood, T., Perelman, E., Calder, B.: Basic block distribution analysis to find periodic behavior and simulation points in applications. In: International Conference on Parallel Architectures and … (January 1991)Google Scholar
- 10.Snavely, A., Carrington, L., Wolter, N., Labarta, J.: A framework for performance modeling and prediction. Supercomputing (January 2002)Google Scholar
- 12.Vetter, J.: Performance analysis of distributed applications using automatic classification of communication inefficiencies. In: ICS 2000: Proceedings of the 14th International Conference on Supercomputing, New York, NY, USA, pp. 245–254 (2000)Google Scholar
- 13.Wong, A., Rexachs, D., Luque, E.: Parallel application signature. In: IEEE International Conference on Cluster Computing and Workshops, CLUSTER 2009, August 31-September 4, pp. 1–4 (2009)Google Scholar
- 14.Wong, A., Rexachs, D., Luque, E.: Parallel application signature for performance prediction. In: International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA 2010), vol. 2. CSREA Press (2010)Google Scholar