Analysis of GPGPU Platforms Efficiency in General-Purpose Computations

Proceedings of the 2011 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA'2011) 2011
Pavel Kartashev, Vladislavs Nazaruks

Nowadays a technique of using graphics processing units (GPUs) for general-purpose computing (or GPGPU) is becoming more and more widespread. The goal of this paper is to analyze efficiency of computing with use of the GPGPU technique, depending on several factors. In this paper, there are analyzed differences in performance and platform organization between widespread GPGPU computational platforms (both hardware and software). There are also described differences between CPU and GPU computations, as well as presented performance measurements for some GPGPU hardware architectures. This paper can help software developers choose more appropriate ways to implement specific fairly large computational tasks.

Atslēgas vārdi
graphics processing units (GPUs), general-purpose computing on graphics processing units (GPGPU), OpenCL
Hipersaite
http://www.cs.uga.edu/~hra/2011-proceedings/pdpta/papers.pdf

Kartashev, P., Nazaruks, V. Analysis of GPGPU Platforms Efficiency in General-Purpose Computations. No: Proceedings of the 2011 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA'2011), Amerikas savienotās valstis, Las Vegas, Nevada, 18.-21. jūlijs, 2011. The United States of America: CSREA Press, 2011, 857.-863.lpp. ISBN 1-60132-193-7.

Publikācijas valoda
English (en)