A parallel computing method using blocked format with optimal partitioning for SpMV on GPU

DOI10.1016/j.jcss.2017.09.010zbMath1383.65041OpenAlexW2766501636WikidataQ126082539 ScholiaQ126082539MaRDI QIDQ1678174

Wangdong Yang, Keqin Li, KenLi Li

Publication date: 14 November 2017

Published in: Journal of Computer and System Sciences (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.jcss.2017.09.010

zbMATH Keywords

partitioningdynamic programmingreorderingsparse matrix-vector multiplicationblocked formatCPU/GPUheterogeneous parallel computinglarge-scale sparse matrices

Mathematics Subject Classification ID

Computational methods for sparse matrices (65F50) Parallel numerical computation (65Y05)

Uses Software

• CUDA
• METIS
• SparseMatrix
• CUBLAS
• MKL
• CUSPARSE
• Sparsity

Cites Work

• Sparse matrix-vector multiplication on the single-chip cloud computer many-core processor
• Towards dense linear algebra for hybrid GPU accelerated manycore systems
• Performance Optimization Using Partitioned SpMV on GPUs and Multicore CPUs
• The university of Florida sparse matrix collection
• An Approximate Minimum Degree Ordering Algorithm
• Improved Three-Way Split Formulas for Binary Polynomial and Toeplitz Matrix Vector Products
• Multiway Splitting Method for Toeplitz Matrix Vector Product
• Nested Dissection of a Regular Finite Element Mesh

This page was built for publication: A parallel computing method using blocked format with optimal partitioning for SpMV on GPU