Welcome to the InfoSci Platform
Reference Hub1
Optimization of Sparse Distributed Computations

Optimization of Sparse Distributed Computations

Olfa Hamdi Larbi
Copyright: © 2022 |Volume: 14 |Issue: 1 |Pages: 18
ISSN: 1938-0259|EISSN: 1938-0267|EISBN13: 9781683180524|DOI: 10.4018/IJGHPC.301586
Cite Article Cite Article

MLA

Larbi, Olfa Hamdi. "Optimization of Sparse Distributed Computations." IJGHPC vol.14, no.1 2022: pp.1-18. http://doi.org/10.4018/IJGHPC.301586

APA

Larbi, O. H. (2022). Optimization of Sparse Distributed Computations. International Journal of Grid and High Performance Computing (IJGHPC), 14(1), 1-18. http://doi.org/10.4018/IJGHPC.301586

Chicago

Larbi, Olfa Hamdi. "Optimization of Sparse Distributed Computations," International Journal of Grid and High Performance Computing (IJGHPC) 14, no.1: 1-18. http://doi.org/10.4018/IJGHPC.301586

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

We address the problem of the optimization of sparse matrix-vector product (SpMV) on homogeneous distributed systems. For this purpose, we propose three approaches based on partitioning the matrix into row blocks. These blocks are defined by a set of a fixed number of rows and a set of contiguous (resp. non-contiguous) rows containing a fixed number of non-zero elements. These approaches lead to solve some specific NP-hard scheduling problems. Thus, adequate heuristics are designed. We analyse the theoretical performance of the proposed approaches and validate them by a series of experiments. This work represents an important step in an overall objective which is to determine the best-balanced distribution for the SpMV computation on a distributed system. In order to validate our approaches for sparse matrix distribution, we compare them to hypergraph model as well as to PETSc library for SpMV distribution on a homogenous multicore cluster. Experimentations show that our approaches provide performances 2 times better than hypergraph and 49 times better than PETSc.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.