Performance analysis of a hybrid overset multi-block application on multiple architectures

This paper presents a detailed performance analysis of a multi-block overset grid CFD application on multiple state-of-the-art computer architectures. The application is implemented using a hybrid MPI+OpenMP programming paradigm that exploits both coarse and fine-grain parallelism. The hybrid model also extends the applicability of multi-block programs to large clusters of SMP nodes. Extensive investigations were conducted on Cray SX-6, IBM Power3 and Power4, and SGI Origin3000 platforms. Overall results for complex vortex dynamics simulations demonstrate that the SX-6 achieves the highest performance and outperforms the RISC-based architectures; however, the best scaling performance was achieved on the Power3.