Applying Process Migration on a BSP-Based LU Decomposition Application
Process migration is an useful mechanism to offer load balancing. In this context, we developed a model called MigBSP that controls processes rescheduling on BSP applications. MigBSP is especially pertinent to obtain performance on this type of applications, since they are composed by supersteps which always wait for the slowest process. In this paper, we focus on the BSP-based modeling of the widely used LU Decomposition algorithm as well as its execution with MigBSP. The use of multiple metrics to decide migrations and adaptations on rescheduling frequency turn possible gains up to 19% over our cluster-of-clusters architecture. Finally, our final idea is to show the possibility to get performance in LU effortlessly by using novel migration algorithms.
KeywordsLoad Balance High Performance Computing Process Migration Migration Cost Application Execution
Unable to display preview. Download preview PDF.
- 2.Bhandarkar, M.A., Brunner, R., Kale, L.V.: Run-time support for adaptive load balancing. In: IPDPS 2000: Proceedings of the 15 IPDPS 2000 Workshops on Parallel and Distributed Processing, pp. 1152–1159. Springer, London (2000)Google Scholar
- 7.Chen, L., Wang, C.-L., Lau, F.: Process reassignment with reduced migration cost in grid load rebalancing. In: IEEE International Symposium on Parallel and Distributed Processing, IPDPS 2008, pp. 1–13 (April 2008)Google Scholar
- 9.da Rosa Righi, R., Pilla, L.L., Carissimi, A., Navaux, P., Heiss, H.-U.: Migbsp: A novel migration model for bulk-synchronous parallel processes rescheduling. In: 10th IEEE International Conference on High Performance Computing and Communications, pp. 585–590 (2009)Google Scholar
- 13.Huang, C., Zheng, G., Kalé, L., Kumar, S.: Performance evaluation of adaptive mpi. In: PPoPP 2006: Proceedings of the Eleventh ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, pp. 12–21. ACM Press, New York (2006)Google Scholar
- 16.Maassen, J., van Nieuwpoort, R.V., Kielmann, T., Verstoep, K., den Burger, M.: Middleware adaptation with the delphoi service. Concurrency and Computation: Practice & Experience (2006)Google Scholar
- 17.Pontelli, E., Le, H.V., Son, T.C.: An investigation in parallel execution of answer set programs on distributed memory platforms: Task sharing and dynamic scheduling. Comput. Lang. Syst. Struct. 36(2), 158–202 (2010)Google Scholar
- 19.Silva, R.E., Pezzi, G., Maillard, N., Diverio, T.: Automatic data-flow graph generation of mpi programs. In: SBAC-PAD 2005: Proceedings of the 17th International Symposium on Computer Architecture on High Performance Computing, Washington, DC, USA, pp. 93–100. IEEE Computer Society, Los Alamitos (2005)Google Scholar
- 21.Wieczorek, M., Podlipnig, S., Prodan, R., Fahringer, T.: Bi-criteria scheduling of scientific workflows for the grid. ccgrid, 9–16 (2008)Google Scholar