摘要：Nowadays, all most personal computers have multi-core processors. We try to exploit computational power from the multi-core architecture. We need a new design on existing algorithms and software. In this paper, we propose the parallelization of the well-known k-means clustering algorithm. We employ Parallel for-Loops (parfor) in MATLAB. Where a loop of n iterations could run on a cluster of m MATLAB workers simultaneously, each worker executes only n/m iterations of the loop. The experimental results demonstrate considerable speedup rate of the proposed parallel k-means clustering method run on a multicore/multiprocessor machine, compared to the serial k-means approach.
关键词：Clustering Algorithms; High Performance Computing; K-means Algorithm; and Parallel Computing
