Efficient sparse matrix-vector multiplication on cache-based GPUs
Sparse matrix-vector multiplication is an integral part of many scientific algorithms. Several studies have shown that it is a bandwidth-limited operation on current hardware. On cache-based architectures the main factors that influence performance are spatial locality in accessing the matrix, and t...
Main Authors: |
Reguly István Zoltán Giles M |
---|---|
Format: | Book part |
Published: |
IEEE Communications Society
Piscataway (NJ)
2012
|
Series: | Innovative Parallel Computing (InPar), 2012
|
Subjects: | |
mtmt: | 2724105 |
Online Access: | https://publikacio.ppke.hu/1901 |
Similar Items
-
Improving efficiency and locality of high-order stencil applications on GPUs
by: Schmidt László
Published: (2024) -
Microfluidic particle separation based on multiple compression chains
by: Kurucz Benedek
Published: (2024) -
GPU-related efficient visual information processing approaches
by: Gelencsér-Horváth Anna
Published: (2023) -
Multi-vector policy in the example of Kazakhstan possibilities and risks /
by: Oshan Bayan
Published: (2023) -
Molecular cytogenetic investigations in multiple myeloma
by: Kondor Nóra
Published: (2024)