![]() Our SpMVM implementations leverage the hybrid paradigm to efficiently overlap expensive communications with computations. We also study important features of this implementation and compare with previously reported implementations that do not exploit underlying symmetry. We analyze the performance of recently proposed implementation of the distributed symmetric SpMVM, originally developed for large sparse symmetric matrices arising in ab initio nuclear structure calculations. In this paper, our target systems are high end multi-core architectures and we use messaging passing interface + open multiprocessing hybrid programming model for parallelism. Due to its low arithmetic intensity, several approaches have been proposed in literature to improve its scalability and efficiency in large scale computations. In this article, sparse matrix vector multiply (SpMVM) is an important kernel that frequently arises in high performance computing applications.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |