Tools for High Performance Computing

by Resch, Michael; Keller, Rainer; Himmler, Valentin; Krammer, Bettina; Schulz, Alexander

• ISBN13: 9783540685616
• ISBN10: 3540685618
• Format: Hardcover
• Copyright: 2008-09-03
• Publisher: Springer-Verlag New York Inc

Summary

"With the advent of multi-core processors, implementing parallel programming methods in application development is absolutely necessary in order to achieve good performance. To support application developers in the various tasks involved in this process, several different tools need to be at his or her disposal. This workshop will give the users an overview of the existing tools in the area of integrated development environments for clusters, various parallel debuggers, and new-style performance analysis tools, as well as an update on the state of the art of long-term research tools, which have advanced to an industrial level."--BOOK JACKET.

Table of Contents

Integrated Development Environments
Sun HPC ClusterTools 7+: A Binary Distribution of Open MPI	p. 3
Introduction	p. 3
History	p. 4
Sun-Driven features	p. 5
Sun Product Activity	p. 13
Pros and Cons	p. 15
Future work and conclusions	p. 16
References	p. 17
An Integrated Environment For the Development of Parallel Applications	p. 19
Introduction	p. 19
Challenges	p. 21
Architecture	p. 23
A Simple Case Study	p. 28
Future Directions	p. 31
Conclusion	p. 33
References	p. 34
Debugging MPI Programs on the Grid using g-Eclipse	p. 35
Introduction	p. 35
Related Work	p. 36
Overview of g-Eclipse Approach	p. 37
Remote Builder	p. 38
Grid Application Launchers	p. 39
Trace Viewer	p. 39
Conclusions and Future Work	p. 44
References	p. 44
Parallel Communication and Debugging
Enhanced Memory debugging of MPI-parallel Applications in Open MPI	p. 49
Introduction	p. 49
Overview of Memcheck	p. 50
Design and Implementation	p. 51
Performance Implications	p. 53
Detectable error classes and findings in actual applications	p. 57
Conclusion and future work	p. 59
References	p. 60
MPI Correctness Checking with Marmot	p. 61
Introduction	p. 62
Related Work	p. 62
Design of Marmot	p. 63
Collaboration with other tools	p. 70
Experiences with real Applications	p. 72
How to install and use Marmot	p. 75
Conclusion and Future Work	p. 76
References	p. 76
Memory Debugging in Parallel and Distributed Applications	p. 79
Introduction	p. 79
The Challenges of Memory Debugging in Parallel Development	p. 80
Classifying Memory Errors	p. 80
Detecting Memory Leaks	p. 82
The MemoryScape Debugger	p. 82
MemoryScape Architecture	p. 83
MemoryScape Features	p. 84
MemoryScape Usage Tips	p. 87
MemoryScape User Case Study: SIMULIA Uses MemoryScape to Find and Fix Bugs Quickly	p. 88
Future MemoryScape Product Plans	p. 90
Conclusion	p. 90
Performance Analysis Tools
Sequential Performance Analysis with Callgrind and KCachegrind	p. 93
Introduction	p. 93
Callgrind: a Call-Graph building Online Cache Simulator	p. 97
KCachegrind: Profile Visualization	p. 105
Usage Example	p. 110
Future Development	p. 111
References	p. 113
Improving Cache Utilization Using Acumem VPE	p. 115
Introduction	p. 116
Throughput Study of SPEC CPU 2006	p. 118
First Generation Performance Tools Based on Hardware Counters	p. 120
Enter: The New Performance Tool	p. 122
Utilization Study of the Worst SPEC CPU 2006 Applications	p. 126
Tuning Example: 179.art	p. 128
Tuning Example: Revisiting the Throughput Applications	p. 132
Conclusion	p. 134
References	p. 135
Parallel Performance Analysis Tools
The Vampir Performance Analysis Tool-Set	p. 139
Introduction	p. 139
Performance Analysis via Profiling or Tracing	p. 140
Instrumentation with VampirTrace	p. 141
Run-Time Measurement and Event Recording	p. 144
Trace Visualization with Vampir and VampirServer	p. 148
Related Work	p. 154
Conclusions and Future Work	p. 154
References	p. 155
Usage of the SCALASCA toolset for scalable performance analysis of large-scale parallel applications	p. 157
Introduction	p. 157
Overview	p. 158
Instrumentation and Measurement	p. 159
Trace Analysis	p. 162
Understanding Performance Behavior	p. 164
Outlook	p. 166
References	p. 167
Evolution of a Parallel Performance System	p. 169
Introduction	p. 169
TAU Performance System Design and Architecture	p. 170
TAU Instrumentation	p. 172
TAU Measurement	p. 178
TAU Analysis	p. 183
Conclusion and Future Work	p. 186
References	p. 188
Cray Performance Analysis Tools	p. 191
Introduction	p. 191
The Cray Performance Analysis Tools	p. 192
Conclusions and Future Work	p. 198
References	p. 199
Index	p. 201
Table of Contents provided by Ingram. All Rights Reserved.