Tools for High Performance Computing (eBook)

Proceedings of the 2nd International Workshop on Parallel Tools for High Performance Computing, July 2008, HLRS, Stuttgart

Rainer Keller, Valentin Himmler, Bettina Krammer, Alexander Schulz (Herausgeber)

eBook Download: PDF

2008
XI, 202 Seiten
Springer Berlin (Verlag)
978-3-540-68564-7 (ISBN)

Infos

Developing software for current and especially for future architectures will require knowledge about parallel programming techniques of applications and library p- grammers. Multi-core processors are already available today, and processors with a dozen and more cores are on the horizon. The major driving force in hardware development, the game industry, has - ready shown interest in using parallel programming paradigms, such as OpenMP for further developments. Therefore developers have to be supported in the even more complex task of programming for these new architectures. HLRS has a long-lasting tradition of providing its user community with the most up-to-date software tools. Additionally, important research and development projects are worked on at the center: among the software packages developed are the MPI correctness checker Marmot, the OpenMP validation suite and the M- implementations PACX-MPI and Open MPI. All of these software packages are - ing extended in the context of German and European community research projects, such as ParMA, the InterActive European Grid (I2G) project and the German C- laborative Research Center (Sonderforschungsbereich 716). Furthermore, ind- trial collaborations, i.e. with Intel and Microsoft allow HLRS to get its software production-grade ready. In April 2007, a European project on Parallel Programming for Multi-core - chitectures, in short ParMA was launched, with a major focus on providing and developing tools for parallel programming.

Inhalt

Preface 5
Contents 7
List of Contributors 11
I Integrated Development Environments 12
Sun HPC ClusterTools™ 7+: A Binary Distribution of Open MPI 13
Introduction 13
History 14
Sun-Driven features 15
Sun Product Activity 23
Pros and Cons 25
Future work and conclusions 26
References 27
An Integrated Environment For the Development of Parallel Applications 29
Introduction 29
Challenges 31
Architecture 33
A Simple Case Study 38
Future Directions 41
Conclusion 43
References 44
Debugging MPI Programs on the Grid using g-Eclipse 45
Introduction 45
Related Work 46
Overview of g-Eclipse Approach 47
Remote Builder 48
Grid Application Launchers 49
Trace Viewer 49
Conclusions and Future Work 54
References 54
II Parallel Communication and Debugging 56
Enhanced Memory debugging of MPI-parallel Applications in Open MPI 57
Introduction 57
Overview of Memcheck 58
Design and Implementation 59
Performance Implications 61
Detectable error classes and findings in actual applications 65
Conclusion and future work 67
References 68
MPI Correctness Checking with Marmot 69
Introduction 70
Related Work 70
Design of Marmot 71
Collaboration with other tools 78
Experiences with real Applications 80
How to install and use Marmot 83
Conclusion and Future Work 84
References 84
Memory Debugging in Parallel and Distributed Applications 87
Introduction 87
The Challenges of Memory Debugging in Parallel Development 88
Classifying Memory Errors 88
Detecting Memory Leaks 90
The MemoryScape Debugger 90
MemoryScape Architecture 91
MemoryScape Features 92
MemoryScape Usage Tips 95
MemoryScape User Case Study: SIMULIA Uses MemoryScape to Find and Fix Bugs Quickly 96
Future MemoryScape Product Plans 98
Conclusion 98
III Performance Analysis Tools 99
Sequential Performance Analysis with Callgrind and KCachegrind 100
Introduction 100
Callgrind: a Call-Graph building Online Cache Simulator 104
KCachegrind: Profile Visualization 112
Usage Example 117
Future Development 118
References 120
Improving Cache Utilization Using Acumem VPE 121
Introduction 122
Throughput Study of SPEC CPU 2006 124
First Generation Performance Tools Based on Hardware Counters 126
Enter: The New Performance Tool 128
Utilization Study of the Worst SPEC CPU 2006 Applications 132
Tuning Example: 179.art 134
Tuning Example: Revisiting the Throughput Applications 138
Conclusion 140
References 141
Parallel Performance Analysis Tools 141
The Vampir Performance Analysis Tool-Set 143
Introduction 143
Performance Analysis via Profiling or Tracing 144
Instrumentation with VampirTrace 145
Run-Time Measurement and Event Recording 148
Trace Visualization with Vampir and VampirServer 152
Related Work 158
Conclusions and Future Work 158
References 159
Usage of the SCALASCA toolset for scalable performance analysis of large-scale parallel applications 160
Introduction 160
Overview 161
Instrumentation and Measurement 162
Trace Analysis 165
Understanding Performance Behavior 167
Outlook 169
References 170
Evolution of a Parallel Performance System 171
Introduction 171
TAU Performance System Design and Architecture 172
TAU Instrumentation 174
TAU Measurement 180
TAU Analysis 185
Conclusion and Future Work 188
References 190
Cray Performance Analysis Tools 193
Introduction 193
The Cray Performance Analysis Tools 194
Conclusions and Future Work 200
References 201
Index 202

Produkt-Details

• Erscheint lt. Verlag: 3.6.2008
• Zusatzinfo: XI, 202 p.
• Verlagsort: Berlin
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Netzwerke
• Themenwelt: Mathematik / Informatik ► Informatik ► Software Entwicklung
• Themenwelt: Technik
• Schlagworte: Calculus • Cluster • Communication • Complexity • Debugging • Desktop • development tools for high performance compting • Eclipse • linear optimization • object oriented design • parallel debugging • parallel performance analysis • programming • Scala • support infrastructure for parallel and high performance com • support infrastructure for parallel and high performance computing
• ISBN-10: 3-540-68564-2 / 3540685642
• ISBN-13: 978-3-540-68564-7 / 9783540685647