Automatic Performance Analysis for CRAY T3E
Automatic Performance Analysis for CRAY T3E
One of the reasons why parallel programming is considered to be a difficult task is that users frequently cannot predict the performance impact of implementation decisions prior to program execution. This results in a cycle of incremental performance improvements based on runtime performance data. W...
Saved in:
Personal Name(s): | Gerndt, Michael (Corresponding Author) |
---|---|
Mohr, Bernd / Pantano, Mario / Wolf, Felix | |
Contributing Institute: |
Zentralinstitut für Angewandte Mathematik; ZAM Jülich Supercomputing Center; JSC |
Published in: |
Proceedings of the Seventh Workshop on Compilers for Parallel Computers |
Imprint: |
1998
|
Physical Description: |
69-78 |
Conference: | Seventh Workshop on Compilers for Parallel Computers, Linköping (Sweden), 1998-06-29 - 1998-07-01 |
Document Type: |
Contribution to a book Contribution to a conference proceedings |
Research Program: |
ohne Topic |
Publikationsportal JuSER |
One of the reasons why parallel programming is considered to be a difficult task is that users frequently cannot predict the performance impact of implementation decisions prior to program execution. This results in a cycle of incremental performance improvements based on runtime performance data. While gathering and analyzing performance data is supported by a large number of tools, typically interactive, the task of performance analysis is still too complex for users. This article illustrates this fact based on the current analysis support on CRAY T3E. As a consequence, we are convinced that automatic analysis tools are required to identify frequently occuring and well-defined performance problems automatically. This article describes the novel design of a generic automatic performance analysis environment called KOJAK. Besides its structure we also outline the first component, EARL, a new meta-tool designed and implemented as a programmable interface to calculate more abstract metrics from existing trace files, and to locate complex patterns describing performance problems. |