Parallel relaxed and extrapolated algorithms for computing PageRank

作     者：Arnal, Josep Migallon, Hector Migallon, Violeta Palomino, Juan A. Penades, Jose 

作者机构：Univ Alicante Dept Ciencia Computac & Inteligencia Artificial Alicante 03071 Spain Univ Miguel Hernandez Dept Fis & Arquitectura Comp Alicante 03202 Spain 

出 版 物：《JOURNAL OF SUPERCOMPUTING》 (超高速计算杂志)

年 卷 期：2014年第70卷第2期

页      面：637-648页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Spanish Ministry of Science and Innovation [TIN2011-26254] 

主　　题：PageRank Parallel algorithms Power method Relaxation and extrapolation 

摘      要：In this paper, parallel Relaxed and Extrapolated algorithms based on the Power method for accelerating the PageRank computation are presented. Different parallel implementations of the Power method and the proposed variants are analyzed using different data distribution strategies. The reported experiments show the behavior and effectiveness of the designed algorithms for realistic test data using either OpenMP, MPI or an hybrid OpenMP/MPI approach to exploit the benefits of shared memory inside the nodes of current SMP supercomputers.