Title
Cooperative Multitasking for GPU-Accelerated Grid Systems
Abstract
Exploiting the graphics processing unit (GPU) is useful to obtain higher performance with a less number of host machines in grid systems. One problem in GPU-accelerated grid systems is the lack of efficient multitasking mechanisms. In this paper, we propose a cooperative multitasking method capable of simultaneous execution of a graphics application and a CUDA-based scientific application on a single GPU. To prevent significant performance drop in frame rate, our method (1) divides scientific tasks into smaller subtasks and (2) serially executes them at the appropriate intervals. Experimental results show that the proposed method is useful to control the frame rate of the graphics application and the throughput of the scientific application. For example, matrix multiplication can be processed at 50% of the dedicated throughput while achieving interactive rendering at 54 frames per second.
Year
DOI
Venue
2012
10.1109/CCGRID.2010.18
Concurrency and Computation: Practice and Experience
Keywords
Field
DocType
scientific task,gpu-accelerated grid systems,efficient multitasking mechanism,cuda-based scientific application,cooperative multitasking,graphics application,scientific application,gpu-accelerated grid system,cooperative multitasking method,dedicated throughput,frame rate,resource management,multitasking,graphics,grid computing,grid,acceleration,kernel,throughput,information science,parallel processing,multiprogramming,coprocessors,frames per second,groupware,matrix multiplication
Grid computing,CUDA,Computer science,Parallel computing,Frame rate,Computer multitasking,Coprocessor,Graphics processing unit,Human multitasking,Rendering (computer graphics)
Journal
Volume
Issue
ISSN
24
1
1532-0626
ISBN
Citations 
PageRank 
978-1-4244-6987-1
4
0.46
References 
Authors
9
4
Name
Order
Citations
PageRank
Fumihiko Ino131738.63
Akihiro Ogita240.46
Kentaro Oita340.46
Kenichi Hagihara452856.94