An Approach Based on Partial Executions for Evaluating the Scalability of Parallel Applications

• Reilta Christine Dantas Maia UFRN
• Samuel Xavier-de-Souza UFRN

DOI: https://doi.org/10.5753/erad-ne.2025.11751

Abstract

Scalability analysis is crucial for developers to gauge how their applications perform under varying resource and workload conditions. However, analyzing complex applications can be time-consuming. This study proposes a partial execution approach in the Parallel Scalability Analyzer tool, leveraging the Paramount Iteration technique. This approach holds significant potential for reducing scalability analysis time.

References

da Silva, V., da Silva, A., Valderrama, C., Manneback, P., and Xavier-de Souza, S. (2022). A minimally intrusive approach for automatic assessment of parallel performance scalability of shared-memory hpc applications. Electronics, 11:689.

Nguyen, H. T., Wei, L., Bhatele, A., Gamblin, T., Boehme, D., Schulz, M., Ma, K.-L., and Bremer, P.-T. (2016). VIPACT: A visualization interface for analyzing calling context trees. In Proceedings of the Workshop on Visual Performance Analysis, pages 1–8. IEEE Press.

Tavares, W., Reis, L., Brunetta, J., and Borin, E. (2019). Aplicação da técnica paramount iteration nas aplicações blast e dnn-rom na nuvem computacional. In Anais do XX Simpósio em Sistemas Computacionais de Alto Desempenho, pages 228–239. SBC.

Yang, L. T., Ma, X., and Mueller, F. (2005). Cross-platform performance prediction of parallel applications using partial execution. IEEE.