| 摘要: |
| 在对HITS算法和基于MapReduce编程模型的云计算框架Hadoop的研究基础上,利用Hadoop来重新设计并实现HITS算法.同时,在实验中分析了不同blocksize和集群规模对算法执行效率的影响.实验表明：当blocksize过大时,由于没有充分利用集群的并行特性,算法效率逐渐降低,而适当扩大集群规模,算法运行效率会逐渐提高. |
| 关键词: HITS MapReduce Hadoop 分快大小 集群 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| The realization of HITS algorithm based on MapReduce |
|
YU Hui, WANG Xiaomei
|
|
College of Information,Mechanical and Electrical Engineering,Shanghai Normal University
|
| Abstract: |
| In this paper,we use Hadoop to redesign HITS algrothim according to the research of HITS algrothim and the cloud computing framework Hadoop which base on MapReduce progaming model.Meanwhile,we analysize the effect of diffenent blocksize and cluster scale in influent on the efficiency of the algrothim.The results shows that the efficiency reduced with the blocksize increased,while the efficiency increased gradually when cluster scale had been expanded properly. |
| Key words: HITS MapReduce Hadoop blocksize cluster |
