Main Contents

• Preface

  • The Aims
  • The Contents
  • The Audience
  • Acknowledgments

• New Building Blocks for HPC in 1995

  • General Discussion
  • Geographical Distribution
  • Market Shares of Vendors
  • Architectural Changes
  • Technological Changes
  • Application Areas
  • Conclusions
  • References

• High Performance Computing in the U.S. in 1995

  • Introduction
  • U.S. Dominance of the World Wide HPC Market
  • Market enetration by Technology and Architecture
  • Conclusions
  • References

• Developments in Japan

  • Background
  • Current commercial offerings
  • Current market situation
    • Government market
    • General market
    • Market distribution
  • The Japanese TOP20
  • Conclusions and outlook
  • References

• High Performance Computing in Europe

  • General Situation
    • Distribution on country groups
    • The different manufacturers
    • Architectures
  • Germany
  • United Kingdom
  • France
  • Benelux
  • The European TOP25 Supercomputer Centres

• The TOP25 Supercomputer Sites.

  • Introduction
  • List of the TOP25 Sites
  • Vendor Sites
  • Background Information about some of the TOP25 sites
    • National Aerospace Laboratory Numerical Wind Tunnel, Tokyo, Japan - 184.409 Gflop/s
    • Oak Ridge National Laboratory, Oak Ridge, Tennessee - 159.900 Gflop/s
    • National Security Agency, Fort Meade, Maryland - 156.471 Gflop/s
    • Los Alamos National Laboratory, Los Alamos, New Mexico - 155.400 Gflop/s
    • Sandia National Laboratory, Albuquerque, New Mexico - 143.400 Gflop/s
    • University of Minnesota, Minneapolis, - 124.620 Gflop/s
    • National Laboratory for High Energy Physics (KEK), Japan - 98.900 Gflop/s
    • Cornell University, Ithaca, NY - 96.024 Gflop/s
    • University of Tokyo, Japan - 83.776 Gflop/s
    • Maui High Performance Computing Center, Kihei, Hawaii - 83.700 Gflop/s
    • Japan Atomic Energy Institute, Japan - 75.918 Gflop/s
    • Tohuku University, Japan - 71.300 Gflop/s
    • Pittsburgh Supercomputer Center, Pittsburgh, Pennsylvania - 64.500 Gflop/s
    • Atmospheric Env. Serv., Dorval, Canada - 58.670 Gflop/s
    • Nagoya University, Japan - 58.509 Gflop/s
  • References

• Short description of newly featuring archictectures in the Top 500

  • The new list of the TOP500 systems
    • The Convex Exemplar SPP-1200.
    • The Cray Y-MP T90 (Triton)
    • The Intel Paragon XP MP
    • The NEC SX-4
    • Silicon Graphics PowerChallenge Array
  • References

• MPI: A STANDARD MESSAGE PASSING INTERFACE

  • Introduction
  • An Overview of MPI
  • The Communicator Abstraction
  • Point-To-Point Communication
    • Communication Modes
    • Persistent Communication Requests
  • General Datatypes
  • Application Topologies
  • Collective Communication
    • Collective Data Movement Routines
    • Global Computation Routines
  • Summary
  • References

• VAMPIR: Visualization and Analysis

  • Introduction
  • The Message Passing Interface (MPI)
  • The VAMPIR Environment
  • Program Instrumentation
  • Visualization of System Activities
  • Statistics
  • Time-line Displays
  • Additional Features
  • Summary and Conclusions
  • References

• Cluster of Supercomputers - an Alternative to Massively Parallel Computing

  • Introduction
  • Cluster Efficiency
  • Cluster computing assumptions
  • The Parallel IFS - Partitioning Concepts
  • Main result
  • Performance Modelling
  • Discussion of results
  • Summary
  • References