Aspects of GPU for general purpose high performance computing

Author

Suda, Reiji ; Aoki, Takayuki ; Hirasawa, Shoichi ; Nukada, Akira ; Honda, Hiroki ; Matsuoka, Satoshi

Author_Institution

Grad. Sch. of Inf. Sci. & Technol., Univ. of Tokyo, Tokyo

fYear

2009

fDate

19-22 Jan. 2009

Firstpage

216

Lastpage

223

Abstract

We discuss hardware and software aspects of GPGPU, specifically focusing on NVIDIA cards and CUDA, from the viewpoints of parallel computing. The major weak points of GPU against newest supercomputers are identified to be and summarized as only four points: large SIMD vector length, small memory, absence of fast L2 cache, and high register spill penalty. As software concerns, we derive optimal scheduling algorithm for latency hiding of host-device data transfer, and discuss SPMD parallelism on GPUs.

Keywords

computer graphic equipment; computer graphics; parallel machines; scheduling; specification languages; CUDA programming language; GPGPU system; NVIDIA card; SIMD vector length; SPMD parallelism; general purpose high performance computing; graphic processing unit; host-device data transfer; optimal scheduling algorithm; parallel computing; supercomputer; Computational fluid dynamics; Concurrent computing; Delay; Hardware; High performance computing; Information systems; Parallel processing; Software algorithms; Software performance; Supercomputers;

fLanguage

English

Publisher

ieee

Conference_Titel

Design Automation Conference, 2009. ASP-DAC 2009. Asia and South Pacific

Conference_Location

Yokohama

Print_ISBN

978-1-4244-2748-2

Electronic_ISBN

978-1-4244-2749-9

Type

conf

DOI

10.1109/ASPDAC.2009.4796483

Filename

4796483