Performance Prediction of Oblivious BSP Programs

The BSP model can be extended with a zero cost synchronization mechanism, which can be used when the number of messages due to receives is known. This mechanism, usually known as "oblivious synchronization" implies that different processors can be in different supersteps at the same time. An unwanted consequence of this software improvement is a loss of accuracy in prediction. This paper proposes an extension of the BSP complexity model to deal with oblivious barriers and shows its accuracy.

[1]  Yi-Bing Lin,et al.  Mobility management for cellular telephony networks , 1996, IEEE Parallel & Distributed Technology: Systems & Applications.

[2]  Gheith A. Abandah,et al.  Modeling the communication performance of the IBM SP2 , 1996, Proceedings of International Conference on Parallel Processing.

[3]  Olaf Bonorden,et al.  The Paderborn university BSP (PUB) library-design, implementation and performance , 1999, Proceedings 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing. IPPS/SPDP 1999.

[4]  Agustin Arruabarrena,et al.  Parallel architectures: Assessing the performance of the new IBM SP2 communication subsystem , 1996, IEEE Parallel & Distributed Technology: Systems & Applications.

[5]  Andrea Zavanella,et al.  Predictability of bulk synchronous programs using MPI , 2000, Proceedings 8th Euromicro Workshop on Parallel and Distributed Processing.

[6]  Francisco Almeida,et al.  Design of parallel algorithms for the single resource allocation problem , 2000, Eur. J. Oper. Res..

[7]  Torsten Suel,et al.  BSPlib: The BSP programming library , 1998, Parallel Comput..

[8]  Leslie G. Valiant,et al.  A bridging model for parallel computation , 1990, CACM.