Performance Analysis of Multicore and Multinodal Implementation of SpMV Operation
Beata Bylina, Jarosław Bylina, Przemysław Stpiczyński, Dominik Szałkowski
Citation: Proceedings of the 2014 Federated Conference on Computer Science and Information Systems, M. Ganzha, L. Maciaszek, M. Paprzycki (eds). ACSIS, Vol. 2, pages 569–576 (2014)
Abstract. In this paper the authors present new algorithms for performing sparse matrix-dense vector multiplication (known as SpMV operation). They show parallel version of algorithm, which can be efficiently implemented on the contemporary multicore architectures. Next, they show distributed (so-called multinodal) version targeted at high performance clusters. Both versions are thoroughly tested using different architectures, compiler tools and sparse matrices of different sizes. Considered matrices comes from The University of Florida Sparse Matrix Collection. The performance of the algorithms is compared to the performance of SpMV routine from widely known MKL library.