期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2011
卷号：11
期号：12
页码：29-37
出版社：International Journal of Computer Science and Network Security
摘要：Graphic Processing Unit (GPU), which was traditionally used for image processing, has been widely applied to general computation called GPGPU. Nowadays, a lot of studies using GPUs are progressing and various products are being developed. GPU has many processor cores, and thereby has low power consumption per unit volume. Even several developing environments are already provided, software developing cost remains high, due to the art of programing and a technical knowledge required for the implementation of GPGPU program of the target algorithm ex-ploiting parallelism requires not only realization of the target algorithm, but also knowledge of architecture such as memory hierarchy. In this paper, we propose a framework which enables easy implementation of parallel computing on GPU. This frame-work can popularize GPU programming. We confirm that we are able to do parallel computing on this framework by implementing and evaluating simple genetic algorithms (SGA). We discuss the relationship between computational speed and execution condition.
