Skip to main content

Parallel relaxed and extrapolated algorithms for computing PageRank

Abstract

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 is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

References

  1. Berkhin P (2005) A survey on PageRank computing. Internet Math 2(1):73–120

    Article  MathSciNet  MATH  Google Scholar 

  2. Boldi P, Codenotti B, Santini M, Vigna S (2004) Ubicrawler: a scalable fully distributed Web crawler. Softw Pract Exp 34:711–726

    Article  Google Scholar 

  3. Dongarra J, Huss-Lederman S, Otto S, Snir M, Walkel D (1996) MPI: the complete reference. The MIT Press, Cambridge

    Google Scholar 

  4. Gleich D, Gray A, Greif C, Lau T (2010) An inner-outer iteration for computing PageRank. SIAM J Sci Comput 32(1):349–371

    Article  MathSciNet  MATH  Google Scholar 

  5. Gleich D, Zhukov L, Berkhin P (2005) Fast parallel PageRank: a linear system approach. In: The fourteenth international world wide web conference. ACM Press, New York

  6. Kamvar SD (2010) Numerical algorithms for personalized search in self-organizing information networks. Princeton University Press, New Jersey

    MATH  Google Scholar 

  7. Kamvar SD, Haveliwala TH, Golub GH (2004) Adaptive methods for the computation of PageRank. Linear Algebra Appl 386:51–65

    Article  MathSciNet  MATH  Google Scholar 

  8. Kamvar SD, Haveliwala TH, Manning CD, Golub GH (2003) Exploiting the block structure of the web for computing PageRank. Stanford University Technical, Report, SCCM-03-02

  9. Kamvar SD, Haveliwala TH, Manning CD, Golub GH (2003) Extrapolation methods for accelerating PageRank computations. In: Twelfth international world wide web conference, pp 261–270

  10. Migallón H, Migallón V, Palomino JA, Penadés J (2010) Parallelization strategies for computing PageRank. In: Proceedings of the seventh international conference on engineering computational technology. Civil-Comp Press, Stirlingshire, Paper 29. doi:10.4203/ccp.94.29

  11. OpenMP official site (2008) http://www.openmp.org

  12. Page L, Brin S, Motwani R, Winograd T (1999) The PageRank citation ranking: Bringing order to the Web. Technical Report, Stanford Digital Library Technologies Project

  13. Rungsawang A, Manaskasemsak B (2012) Fast PageRank computation on a GPU cluster. In: 20th Euromicro international conference on parallel, distributed and network-based processing, pp 450–456

  14. Rungsawang A, Manaskasemsak B (2006) Parallel adaptive technique for computing PageRank. In: Proceedings of the 14th euromicro international conference on parallel, distributed, and network-based processing, PDP’06, pp 15–20

  15. Wilkinson JH (1998) The algebraic eigenvalue problem. Oxford University Press, Oxford

    Google Scholar 

  16. Wu G, Wei Y (2010) An Arnoldi-Extrapolation algorithm for computing PageRank. J Comput Appl Math 234:3196–3212

    Article  MathSciNet  MATH  Google Scholar 

  17. Zhang H, Goel A, Govindan R, Mason K, Van Roy B (2004) Making Eigenvector-based reputation systems robust to collusion. Lect Notes Comput Sci 3243:92–104

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Violeta Migallón.

Additional information

This research was partially supported by the Spanish Ministry of Science and Innovation under Grant Number TIN2011-26254.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Arnal, J., Migallón, H., Migallón, V. et al. Parallel relaxed and extrapolated algorithms for computing PageRank. J Supercomput 70, 637–648 (2014). https://doi.org/10.1007/s11227-014-1118-9

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11227-014-1118-9

Keywords

  • PageRank
  • Parallel algorithms
  • Power method
  • Relaxation and extrapolation