CUKNN: A parallel implementation of K-nearest neighbor on CUDA-enabled GPU

Recent development in Graphics Processing Units (GPUs) has enabled inexpensive high performance computing for general-purpose applications. Due to GPU´s tremendous computing capability, it has emerged as the co-processor of the CPU to achieve a high overall throughput. CUDA programming model provides the programmers adequate C language like APIs to better exploit the parallel power of the GPU. K-nearest neighbor is a widely used classification technique and has significant applications in various domains. The computational-intensive nature of KNN requires a high performance implementation. In this paper, we present a CUDA-based parallel implementation of KNN, CUKNN, using CUDA multi-thread model. Various CUDA optimization techniques are applied to maximize the utilization of the GPU. CUKNN outperforms significantly and achieve up to 15.2X speedup. It also shows good scalability when varying the dimension of the training dataset and the number of records in training dataset.