The Past, Present, and Future of GPU-Accelerated Grid Computing

The emergence of compute unified device architecture (CUDA), which relieved application developers from understanding complex graphics pipelines, made the graphics processing unit (GPU) useful not only for graphics applications but also for general applications. In this paper, we introduce a cycle sharing system named GPU grid, which exploits idle GPU cycles for acceleration of scientific applications. Our cycle sharing system implements a cooperative multitasking technique, which is useful to execute a guest application remotely on a donated host machine without causing a significant slowdown on the host machine. Because our system has been developed since the pre-CUDA era, we also present how the evolution of GPU architectures influenced our system.