The Chunks and Tasks Matrix Library

We present a C++ header-only parallel sparse matrix library, based on sparse quadtree representation of matrices using the Chunks and Tasks programming model. The library implements a number of sparse matrix algorithms for distributed memory parallelization that are able to dynamically exploit data locality to avoid movement of data. This is demonstrated for the example of block-sparse matrix-matrix multiplication applied to three sequences of matrices with different nonzero structure, using the CHT-MPI 2.0 runtime library implementation of the Chunks and Tasks model. The runtime library succeeds to dynamically load balance the calculation regardless of the sparsity structure. (C) 2022 The Authors. Published by Elsevier B.V.

Automated summary

We introduce a C++ header-only parallel sparse matrix library based on a sparse quadtree representation and the Chunks and Tasks programming model. This library dynamically exploits data locality to avoid data movement and demonstrates successful distributed memory parallelization of block-sparse matrix-matrix multiplication for matrices with different nonzero structures.