Skip to Main content Skip to Navigation
New interface
Journal articles

Exploiting Multiple Levels of Parallelism in Sparse Matrix-Matrix Multiplication

Abstract : Sparse matrix-matrix multiplication (or SpGEMM) is a key primitive for many high-performance graph algorithms as well as for some linear solvers, such as algebraic multigrid. The scaling of existing parallel implementations of SpGEMM is heavily bound by communication. Even though 3D (or 2.5D) algorithms have been proposed and theoretically analyzed in the flat MPI model on Erdös--Rényi matrices, those algorithms had not been implemented in practice and their complexities had not been analyzed for the general case. In this work, we present the first implementation of the 3D SpGEMM formulation that exploits multiple (intranode and internode) levels of parallelism, achieving significant speedups over the state-of-the-art publicly available codes at all levels of concurrencies. We extensively evaluate our implementation and identify bottlenecks that should be subject to further research. Read More:
Document type :
Journal articles
Complete list of metadata
Contributor : Laura Grigori Connect in order to contact the contributor
Submitted on : Wednesday, January 4, 2017 - 12:46:25 PM
Last modification on : Tuesday, November 29, 2022 - 12:12:15 PM

Links full text



Ariful Azad, Grey Ballard, Aydin Buluc, James W. Demmel, Laura Grigori, et al.. Exploiting Multiple Levels of Parallelism in Sparse Matrix-Matrix Multiplication. SIAM Journal on Scientific Computing, 2016, 38 (6), pp.27. ⟨10.1137/15M104253X⟩. ⟨hal-01426294⟩



Record views