With the extensive application of network, and the data of network grows rapidly, the process of big data faces severe challenge.After studying the AP clustering algorithm, we improved the AP clustering algorithm based on MapReduce model and designed a distributed AP clustering algorithm based on MapReduce in the cloud platform of Hadoop.We examined the distribu-ted AP clustering algorithm on different size of graph data.The result shows that the distributed AP clustering algorithm is of good efficiency and speedup time.%随着网络的普遍应用，网络中产生的数据急剧增长，大规模数据处理面临严峻挑战。本文在对AP聚类算法进行研究的基础上，利用MapReduce编程模型思想对AP聚类算法进行改进，设计在云平台Hadoop环境下运行的基于MapRe-duce的分布式AP聚类算法，并在实验中对不同规模的图数据进行聚类测试，实验结果表明分布式的AP聚类算法具有很好的时间效率和加速比。
展开▼
