Parallel sparse matrix-matrix multiplication: a scalable solution with 1D algorithm

  title={Parallel sparse matrix-matrix multiplication: a scalable solution with 1D algorithm},
  author={Mohammad Asadul Hoque and Md Rezaul Karim Raju and Christopher John Tymczak and Daniel Vrinceanu and Kiran Chilakamarri},
This paper presents a novel implementation of parallel sparse matrix-matrix multiplication using distributed memory systems on heterogeneous hardware architecture. The proposed algorithm is expected to be linearly scalable up to several thousands of processors for matrices with dimensions over 10 (million). Our approach of parallelism is based on 1D decomposition and can work for both structured and unstructured sparse matrices. The storage mechanism is based on distributed hash lists, which… CONTINUE READING


Publications referenced by this paper.
Showing 1-10 of 21 references

Direct Methods for Sparse Linear Systems. Society for Industrial and Applied Mathematics, Philadelphia

  • Timothy A. Davis
  • PA, USA,
  • 2006
1 Excerpt

Similar Papers

Loading similar papers…