Scientific Supercomputing: Table of Contents

1. Introduction

2. Prototypes with their Bottlenecks

3. Arithmetic Operations and Memory Bandwidth

4. The CRAY T90

5. The Fujitsu VPP300 (and VPP700)

6. The IBM RS/6000 SP

7. The CRAY T3E and DEC Alpha Processors

8. Performance Analysis

9. Basic Considerations Concerning Data Structures

10. Fortran, Autovectorization and Autoparallelization, Programming Models

11. Recurrences

12. Matrix Operations

13. Systems of Linear Equations on Vector Processors

14. Remarks to the Parallelization of the Gauss Algorithm

15. Tridiagonal Linear Systems

16. Iterative Solution of Linear Systems of Equations

17. Explaining the Gap Between Theoretical Peak Performance and Real Performance for Supercomputer Architectures

18. References

End

Addendum