A parallel design of computer Go engine on CUDA-enabled GPU

With the rapid growth of Graphics Processing Unit (GPU) processing capability, using GPU as a coprocessor to assist the CPU in parallel computing has become indispensable. CUDA (Compute Unified Device Architecture) programming model also gives C/C++ language support which makes programming easily. This paper details how to design an engine of computer Go with Monte-Carlo algorithm which is based on GPU with Fermi architecture. We analyze the characteristics of Monte-Carlo algorithm, combined with the CUDA architecture features, divide the algorithm into various sub-modules for GPU computing fast and easily.