Skip to main navigation Skip to search Skip to main content

Early evaluation of the IBM p690

  • P. H. Worley
  • , T. H. Dunigan
  • , M. R. Fahey
  • , J. B. White
  • , A. S. Bland

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Oak Ridge National Laboratory recently received 27 32-way IBM pSeries 690 SMP nodes. In this paper, we describe our initial evaluation of the p690 architecture, focusing on the performance of benchmarks and applications that are representative of the expected production workload.

Original languageEnglish
Title of host publicationProceedings of the IEEE/ACM SC 2002 Conference, SC 2002
PublisherAssociation for Computing Machinery
ISBN (Electronic)076951524X
DOIs
StatePublished - Nov 16 2002
Event2002 IEEE/ACM Conference on Supercomputing, SC 2002 - Baltimore, United States
Duration: Nov 16 2002Nov 22 2002

Publication series

NameProceedings of the International Conference on Supercomputing
Volume2002-November

Conference

Conference2002 IEEE/ACM Conference on Supercomputing, SC 2002
Country/TerritoryUnited States
CityBaltimore
Period11/16/0211/22/02

Funding

∗This research was sponsored by the Office of Mathematical, Information, and Computational Sciences, Office of Science, U.S. Department of Energ y under Contract No. DE-AC05-00OR22725 with UT-Batelle, LLC. According ly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes. †Computer Science and Mathematics Division, Oak Ridge National Laboratory, P.O. Box 2008, Bldg. 6012, Oak Ridge, TN 37831-6367 ([email protected]) ‡Computer Science and Mathematics Division, Oak Ridge National Laboratory, P.O. Box 2008, Bldg. 6012, Oak Ridge, TN 37831-6367 ([email protected]) §Computer Science and Mathematics Division, Oak Ridge National Laboratory, P.O. Box 2008, Bldg. 4500N, Oak Ridge, TN 37831-6203 ([email protected] ov) ¶Computer Science and Mathematics Division, Oak Ridge National Laboratory, P.O. Box 2008, Bldg. 4500N, Oak Ridge, TN 37831-6203 ([email protected]) ‖Computer Science and Mathematics Division, Oak Ridge National Laboratory, P.O. Box 2008, Bldg. 4500N, Oak Ridge, TN 37831-6203 ([email protected]) This research was sponsored by the Office of Mathematical, Information, and Computational Sciences, Office of Science, U.S. Department of Energy under Contract No. DE-AC05-00OR22725 with UT-Batelle, LLC. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes. We gratefully acknowledge the following institutions for access to their HPC systems: • the Pittsburgh Supercomputing Center for access to the Compaq AlphaServer SC at PSC, • the Center for Computational Sciences at ORNL for access to the IBM p690, IBM SP, and Compaq AlphaServer SC at ORNL, and • the National Energy Research Scientific Computing Center for access to the IBM SP and the Cray Research T3E at NERSC, and the aid and encouragement of the following computational scientists and developers: • Warren Washington and Tom Bettge at the National Center for Atmospheric Research for specifying and providing the PCTM benchmark, • John Blondin at North Carolina State University, Tony Mezzacappa at ORNL, and William (Raph) Hix at ORNL for specifying and providing the EVH1 benchmark, • Don Batchelor and Fred Jaeger at ORNL for specifying and providing the AORSA3D benchmark, and • Vance Shaffer and Farid Parpia of IBM for supplying the process binding scripts and libraries.

Cite this