Evaluating Sparse Matrix-Vector Product on the FinisTerrae Supercomputer

In this paper the sparse matrix-vector product (SpMV) is evaluated on the FinisTerrae supercomputer. FinisTerrae is a SMP-NUMA system with more than 2500 processors. Several topics are studied. We have estimated the influence of data and thread allocation in the SpMV performance. Due to the indirect and irregular memory access patterns of SpMV, we have also studied the influence of the memory hierarchy in the performance. Additionally, different parallelization strategies and compiler options were considered. According to the behavior observed in the study a set of optimizations specially tuned for FinisTerrae were successfully applied to SpMV. Noticeable improvements are obtained in comparison with the SpMV na¨ıve implementation.

keywords: sparse matrix, NUMA, data locality, performance