Evaluating Performance and Scalability of the Sparse Linear Systems Solver Spliss

Abstract

Solving linear equation systems is an integral part of implicit methods in computational fluid dynamics (CFD). The sparse linear system solver Spliss aims to provide a linear solver library that, on the one hand, is tailored to requirements of CFD applications but, on the other hand, independent of the particular CFD solver. Focusing on the specific task of solving linear systems allows for integrating more advanced, but also more complex, hardware-specific optimizations, while at the same time hiding this complexity from the CFD solver. Spliss enables the execution of the computationally intensive linear solver on GPUs without the necessity of any code adaption. This work evaluates performance and scalability of Spliss using CODA as an example. CODA is the CFD software being developed as part of a collaboration between the French Aerospace Lab ONERA, the German Aerospace Center (DLR), Airbus, and their European research partners. CODA is jointly owned by ONERA, DLR and Airbus. The evaluation includes an assessment of the scalability on an HPC system based on the AMD Naples architecture and demonstrates the seamless integration of GPUs on a cluster based on Nvidia V100 GPUs.