Abstract | ||
---|---|---|
In recent times, with the advancing of the graphics processing unit (GPU), parallel computing using general-purpose computing on GPU (GPGPU) is expanding. This is achieved through a processing speed faster than those of traditional computing environments across many fields, such as science, medicine, engineering, and analysis. However, there are many constraints in implementing a parallel program using the GPU technology. In this study, we port a CPU-based program (video quality measurement program) to use the technology. Furthermore, we study the acceleration of the GPU-based program and discuss the technical constraints and problems that occur when you modify the CPU-based program to GPU-based program. The program ported to GPU-based program shows more than twice the execution speed improvement rate than the CPU-based program. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1145/2987386.2987414 | RACS |
Keywords | Field | DocType |
GPGPU, Parallel Computing, CUDA | Central processing unit,Computer science,CUDA,Parallel computing,Computational science,General-purpose computing on graphics processing units,Acceleration,Porting,Graphics processing unit,Video quality | Conference |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Seonguk Lee | 1 | 0 | 0.68 |
Jeonghwan Lee | 2 | 0 | 1.01 |
Kyoungmin Kim | 3 | 6 | 2.10 |
Joonhyouk Jang | 4 | 0 | 1.35 |