|
This article is cited in 1 scientific paper (total in 1 paper)
Computer Science, Engineering and Control
Estimate of locality of parallel algorithms implemented on GPUs
N. A. Likhoded, M. A. Paliashchuk Belarusian State University (Nezavisimosti avenue 4, Minsk, 220030 Republic of Belarus)
Abstract:
The problem of obtaining blocks of operations and threads of parallel algorithm resulting in a smaller number of accesses to global memory and resulting in the efficient use of caches and shared memory graphics processor is investigated. We formulated and proved statements to assess the volume of communication transactions generated by alternative sizing of blocks, as well as to minimize the number of cache misses due to the use of temporal and spatial locality of data. The research is constructive and allows software implementation for practical use.
Keywords:
parallel computing, GPU, minimization of communications, temporal locality, spatial locality.
Received: 02.03.2016
Citation:
N. A. Likhoded, M. A. Paliashchuk, “Estimate of locality of parallel algorithms implemented on GPUs”, Vestn. YuUrGU. Ser. Vych. Matem. Inform., 5:3 (2016), 96–111
Linking options:
https://www.mathnet.ru/eng/vyurv147 https://www.mathnet.ru/eng/vyurv/v5/i3/p96
|
Statistics & downloads: |
Abstract page: | 132 | Full-text PDF : | 45 | References: | 22 |
|