1st Wksp on Productivity & Performance (PROPER 2008)
at Euro-Par 2008 Conference, Las Palmas de Gran Canaria, Spain, 25 August 2008
The current and future increase of computer performance is driven by raising the number of cores on a single chip and within a supercomputer. As a consequence, applications need to harness much higher degrees of parallelism in order to satisfy their growing demands for computing power. Writing code that runs correctly and efficiently on large numbers of processors and cores is extraordinarily challenging. The increased concurrency levels also place higher demands on the development process and thus require adequate and scalable tool support for debugging and performance analysis. This workshop will serve as a forum for discussing methods and experiences related to scalability and usability of HPC programming tools, including integration with compilers and the overall development environment.
- tools and tool approaches for parallel program development and analysis
- correctness checking
- performance measurement and evaluation
- success stories about optimization or parallel scalability achieved using tools
Emphasis will be put on giving younger researchers a chance to present their work. There is a student grant available for up to four student researchers presenting their work. It will cover the conference fee plus a maximum of 600 € for travel expenses. Please indicate whether you want to be considered for the student grant with your submission.
The workshop will take place on Monday, Aug 25th 2008 in conjunction with the EuroPar 2008 conference.
Matthias S. Müller, local chair will be Andreas Knüpfer
9:30 - 11:00
- Invited talk: Martin W. J. Schulz (Lawrence Livermore National Laboratory): Scalable Performance Analysis: From Open|SpeedShop to a Community Infrastructure
- Christian Iwainsky, Dieter an Mey: Comparing the usability of performance analysis tools
11:30 - 13:00
- Kevin A. Huck, Wyatt Spear, Allen D. Malony, Sameer Shende, Alan Morris: Parametric Studies in Eclipse with TAU and PerfExplorer
- Karl Fürlinger, Dan Terpstra, Haihang You, Phil Mucci, and Shirley Moore: Enabling Data Structure Oriented Performance Analysis with Hardware Performance Counter Support
- R. Castillo, F. Corbera, A. Navarro, R. Asenjo, and E.L. Zapata: Complete DefUse-Analysis in recursive programs with dynamic data structures
14:30 - 16:00
- Holger Mickler, Andreas Knüpfer, Michael Kluge, Matthias S. Müller, Wolfgang E. Nagel: Trace-Based Analysis and Optimization for the Semtex CFD Application -- Hidden Remote Memory Accesses and I/O Performance
- Zoltan Szebenyi, Brian J. N. Wylie, Felix Wolf: SCALASCA Parallel Performance Analyses of PEPC