An effective model of CPU/GPU collaborative computing in GPU clusters

Remote procedure call (RPC) is a simple, transparent and useful paradigm for providing communication between two processes across a network. The compute unified device architecture (CUDA) programming toolkit and runtime enhance the programmability of the graphics processing unit (GPU) and make GPU more versatile in high performance computing. The current researches mainly focus on the acceleration of algorithms on a GPU or multiple GPUs on a single host. This paper proposes a CPU/GPU collaborative model which can transparently use remote CPU/GPU computing resources to accelerate the computation. The objective is to efficiently manage CPU/GPU resources in a cluster to achieve load balancing.