Tuning Application in a Multi-cluster Environment

  • Eduardo Argollo
  • Adriana Gaudiani
  • Dolores Rexachs
  • Emilio Luque
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4128)


The joining of geographically distributed heterogeneous clusters of workstations through the Internet can be a simple and effective approach to speed up a parallel application execution. This paper describes a methodology to migrate a parallel application from a single-cluster to a collection of clusters, guaranteeing a minimum level of efficiency. This methodology is applied to a parallel scientific application to use three geographically scattered clusters located in Argentina, Brazil and Spain. Experimental results prove that the speedup and efficiency estimations provided by this methodology are more than 90% precision. Without the tuning process of the application a 45% of the maximum speedup is obtained whereas a 94% of that maximum speedup is attained when a tuning process is applied. In both cases efficiency is over 90%.


Local Cluster Network Throughput Maximum Speedup Parallel Application Tuning Process 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Foster, I.: The grid: A new infrastructure for 21st century science. Physics Today 55(2), 42–47 (2002)CrossRefGoogle Scholar
  2. 2.
    Beaumont, O., Legrand, A., Robert, Y.: The master-slave paradigm with heterogeneous processors. IEEE Trans. Parallel Distributed Systems 14(9), 897–908 (2003)CrossRefMathSciNetGoogle Scholar
  3. 3.
    Javadi, B., Akbari, M., Abawajy, J.: Performance analysis of heterogeneous multi-cluster systems. In: ICPP 2005, pp. 493–500 (2005)Google Scholar
  4. 4.
    Bal, H.E., Plaat, A., Bakker, M.G., Dozy, P., Hofman, R.F.H.: Optimizing parallel applications for wide-area clusters. In: Proceedings of IPPS/SPDP 1998, pp. 784–790 (1998)Google Scholar
  5. 5.
    Argollo, E., Rexachs, D., Tinetti, F., Luque, E.: Efficient Execution of Scientific Computation on Geographically Distributed Clusters. In: Dongarra, J., Madsen, K., Waśniewski, J. (eds.) PARA 2004. LNCS, vol. 3732, Springer, Heidelberg (2006)CrossRefGoogle Scholar
  6. 6.
    Aida, K., Natsume, W., Futakata, Y.: Distributed computing with hierarchical master-worker paradigm for parallel branch and bound algorithm. In: CCGrid 2003, pp. 156–163 (2003)Google Scholar
  7. 7.
    Nieuwpoort, R.V., Kielmann, T., Bal, H.E.: Efficient load balancing for wide-area divide-and-conquer applications. In: PPoPP 2001, pp. 34–43. ACM Press, New York (2001)CrossRefGoogle Scholar
  8. 8.
    Argollo, E., de Souza, J.R., Rexachs, D., Luque, E.: Efficient Execution on Long-Distance Geographically Distributed Dedicated Clusters. In: Kranzlmüller, D., Kacsuk, P., Dongarra, J. (eds.) EuroPVM/MPI 2004. LNCS, vol. 3241, pp. 311–319. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  9. 9.
    Carusela, M.F., Perazzo, R.P.J., Romanelli, L.: Stochastic resonant memory storage device. Physical Review 64(3 pt 1), 31101 (2001)Google Scholar
  10. 10.
    Colombet, L., Desbat, L.: Speedup and efficiency of large-size applications on heterogeneous networks. Theoretical Computer Science 196, 31–44 (1998)MATHCrossRefMathSciNetGoogle Scholar
  11. 11.
    McNamara, B., Wiesenfeld, K.: Theory of stochastic resonance. Physical Review A 39, 4854–4869 (1989)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Eduardo Argollo
    • 1
  • Adriana Gaudiani
    • 2
  • Dolores Rexachs
    • 1
  • Emilio Luque
    • 1
  1. 1.Computer Architecture and Operating System DepartmentUniversidad Autónoma de BarcelonaBarcelonaSpain
  2. 2.Instituto de Ciências, InformáticaUniversidad Nacional de General SarmientoBuenos AiresArgentina

Personalised recommendations