Measuring Computer Performance

  • William Jalby
  • David C. Wong
  • David J. Kuck
  • Jean-Thomas Acquaviva
  • Jean-Christophe Beyler

Abstract

Computer performance improvement embraces many issues, but is severely hampered by existing approaches that examine one or a few topics at a time. Each problem solved leads to another saturation point and serious problem. In the most frustrating cases, solving some problems exacerbates others and achieves no net performance gain. This paper discusses how to measure a large computational load globally, using as much architectural detail as needed. Besides the traditional goals of sequential and parallel system performance, these methods are useful for energy optimization.

Keywords

Line Choice Decan Dispatch Aliasing Padding 

References

  1. 1.
    Djoudi, L., Barthou, D., Carribault, P., Lemuet, C., Thomas Acquaviva, J., Jalby, W.: MAQAO: Modular assembler quality analyzer and optimizer for Itanium 2. In: Workshop on EPIC, San Jose, USA (2005) Google Scholar
  2. 2.
    Hockney, R.W., Jesshope, C.R.: Parallel Computers. Hilger, Bristol (1981) Google Scholar
  3. 3.
    Koliai, S., Zuckerman, S., Oseret, E., Ivascot, M., Moseley, T., Quang, D., Jalby, W.: A balanced approach to application performance tuning. In: LCPC, pp. 111–125 (2009) Google Scholar
  4. 4.
    Kuck, D.J.: Computational capacity-based codesign of computer systems. In this volume Google Scholar
  5. 5.
    Petit, E., Bodin, F., Papaure, G., Dru, F.: Poster reception—ASTEX: a hot path based thread extractor for distributed memory system on a chip. In: SC, p. 141 (2006) Google Scholar
  6. 6.
    Wong, D., Kuck, D.: Solving for simulation problems and its use in correctness enforcement for optimization problems (2010). Internal document Google Scholar

Copyright information

© Springer-Verlag London Limited 2012

Authors and Affiliations

  • William Jalby
    • 1
  • David C. Wong
    • 2
  • David J. Kuck
    • 2
  • Jean-Thomas Acquaviva
    • 1
  • Jean-Christophe Beyler
    • 1
  1. 1.UVSQ/Exascale Computing ResearchVersaillesFrance
  2. 2.Intel CorporationUrbanaUSA

Personalised recommendations