Advertisement

Scalable Runtime for MPI: Efficiently Building the Communication Infrastructure

  • George Bosilca
  • Thomas Herault
  • Pierre Lemarinier
  • Ala Rezmerita
  • Jack J. Dongarra
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6960)

Abstract

The runtime environment of MPI implementations plays a key role to launch the application, to provide out-of-band communications, enabling I/O forwarding and bootstrapping of the connections of high-speed networks, and to control the correct termination of the parallel application. In order to enable all these roles on a exascale parallel machine, which features hundreds of thousands of computing nodes (each of them featuring thousands of cores), scalability of the runtime environment must be a primary goal.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Angskun, T., Bosilca, G., Vander Zanden, B., Dongarra, J.: Optimal routing in binomial graph networks. In: Eighth International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2007, pp. 363–370 (December 2007)Google Scholar
  2. 2.
    Balaji, P., Buntinas, D., Goodell, D., Gropp, W., Krishna, J., Lusk, E., Thakur, R.: PMI: A scalable parallel process-management interface for extreme-scale systems. In: Keller, R., Gabriel, E., Resch, M., Dongarra, J. (eds.) EuroMPI 2010. LNCS, vol. 6305, pp. 31–41. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  3. 3.
    Bosilca, G., Coti, C., Herault, T., Lemarinier, P., Dongarra, J.: Constructing resiliant communication infrastructure for runtime environments. Advances in Parallel Computing 19, 441–451 (2010), doi:10.3233/978-1-60750-530-3-441Google Scholar
  4. 4.
    Bosilca, G., Herault, T., Rezmerita, A., Dongarra, J.: On scalability for mpi runtime systems. In: IEEE International Conference on Cluster Computing (to appear, 2011)Google Scholar
  5. 5.
    Mathematics, and Computer Science Division, A. N. L. MPICH-2, implementation of MPI 2 standard (2006), http://www-unix.mcs.anl.gov/mpi/mpich2/
  6. 6.
    Mathematics, and Computer Science Division, A. N. L. Hydra process management framework (2009), http://wiki.mcs.anl.gov/mpich2/index.php/HydraProcessManagementFramework
  7. 7.
    Sridhar, J., Koop, M., Perkins, J., Panda, D.: ScELA: Scalable and extensible launching architecture for clusters. In: Sridhar, J., Koop, M., Perkins, J., Panda, D. (eds.) HiPC 2008. LNCS, vol. 5374, pp. 323–335. Springer, Heidelberg (2008)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • George Bosilca
    • 1
  • Thomas Herault
    • 1
  • Pierre Lemarinier
    • 2
  • Ala Rezmerita
    • 3
  • Jack J. Dongarra
    • 1
  1. 1.University of TennesseeKnoxvilleUSA
  2. 2.Université de Rennes 1, IRISAFrance
  3. 3.Grand-Large, INRIA Saclay – Université Paris-SudFrance

Personalised recommendations