Performance Analysis of the CFD Solver CODA - Harnessing Synergies between Application and Performance Tools Developers

Abstract

Accurate prediction of fluid flows remains an important field of research and engineering. To this end, computational fluid dynamics (CFD) is widely employed. Due to their high demands on computational resources CFD applications benefit from HPC systems. Continuous performance analysis and optimization is key to efficient utilization of HPC resources. This paper demonstrates the beneficial cooperation between developers of HPC software and performance tools in the context of the CFD solver CODA and the sparse linear system solver Spliss. We investigate concepts used by CODA/Spliss to achieve high scalability and evaluate their effectiveness with performance analysis tools. This example illustrates profits obtained from a close collaboration between HPC application and tool developers. On the one hand, tools support application developers in analyzing and tuning their applications. On the other hand, feedback and requests of application developers inspire tools enhancements. In this work we highlight these aspects with an extended support of non-blocking collectives in performance tools and emphasize the need for sophisticated tool support of multi-threaded MPI applications.