Parallel relaxed and extrapolated algorithms for computing PageRank
| Contribuinte(s) |
Universidad de Alicante. Departamento de Ciencia de la Computación e Inteligencia Artificial Computación de Altas Prestaciones y Paralelismo (gCAPyP) |
|---|---|
| Data(s) |
04/02/2015
04/02/2015
01/02/2014
|
| Resumo |
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. This research was partially supported by the Spanish Ministry of Science and Innovation under Grant Number TIN2011-26254. |
| Identificador |
The Journal of Supercomputing. 2014, 70(2): 637-648. doi:10.1007/s11227-014-1118-9 0920-8542 (Print) 1573-0484 (Online) http://hdl.handle.net/10045/44546 10.1007/s11227-014-1118-9 |
| Idioma(s) |
eng |
| Publicador |
Springer Science+Business Media New York |
| Relação |
http://dx.doi.org/10.1007/s11227-014-1118-9 |
| Direitos |
The final publication is available at Springer via http://dx.doi.org/10.1007/s11227-014-1118-9 info:eu-repo/semantics/restrictedAccess |
| Palavras-Chave | #PageRank #Parallel algorithms #Power method #Relaxation and extrapolation #Ciencia de la Computación e Inteligencia Artificial |
| Tipo |
info:eu-repo/semantics/article |
