Workshop on Extreme-scale Performance Tools
Date
November 16, 2012
Location
in conjunction with
Supercomputing 2012
Salt Lake City, Utah
As we approach exascale, the rising architectural complexity coupled with severe resource limitations with respect to power, memory and I/O, makes performance optimization more critical than ever before. However, all the challenges of scalability, heterogeneity, and resilience that application and system developers face will also affect the development of the tool environment needed to achieve performance objectives. This workshop will serve as a forum for application, system, and tool developers to discuss the requirements of future exascale-enabled performance tools and the roadblocks that need to be addressed on the way.
The workshop is organized by the Virtual Institute - High Productivity Supercomputing, an international initiative of publicly-funded HPC programming-tool builders aimed at the enhancement, integration, and deployment of their products. The event will focus not only on technical issues but also on the community-building process necessary to create an integrated performance-tool suite ready for an international exascale software stack.
Agenda
| 08:30 - 08:45 | Welcome and introduction to VI-HPS [PDF] Felix Wolf, German Research School for Simulation Sciences |
| 08:45 - 09:30 | Keynote Talk: Moving towards exascale via optimization for application- and hardware-aware execution [PDF] Laura Carrington, San Diego Supercomputing Center |
| 09:30 - 10:00 | Unified Performance Data Collection with Score-P [PDF] Bert Wesarg, TU Dresden |
| 10:00 - 10:30 | Coffee break |
| 10:30 - 10:45 | Detecting and analyzing application structure with clustering [PDF] Judith Gimenez, Barcelona Supercomputing Center |
| 10:45 - 11:00 | Scalability Improvements in the TAU Performance System [PDF] Sameer Shende, University of Oregon |
| 11:00 - 11:15 | Scaling lessons in Scalasca [PDF] Brian Wylie, Jülich Supercomputing Centre |
| 11:15 - 11:30 | Single-node optimization: still important [PDF] Bettina Krammer, Université de Versailles St-Quentin-en-Yvelines |
| 11:30 - 11:45 | Application-driven MPI profiling with P^nMPI and mpiP [PDF] Martin Schulz, Lawrence Livermore National Laboratory |
| 11:45 - 12:00 | Beyond Performance Analysis: Automatic Program Tuning [PDF] Michael Gerndt, TU Munich |
| 12:00 - 12:15 | Combined Performance and Power Consumption Modeling and
Optimization with MuMMI [PDF] Shirley Moore, University of Texas at El Paso |
| 12:15 - 12:30 | Discussion |
Time and location
08:30 a.m. to 12:30 p.m., Room: 155-F
Please check the SC'12 workshop page for up-to-date information on the location.
Registration
Please register through the official registration at Supercomputing '12.
Additional Information
Workshop page at Supercomputing '12
Organizing Committee
Felix Wolf, German Research School for Simulation Sciences, Germany
Michael Gerndt, TU Munich, Germany
Bettina Krammer, Université de Versailles St-Quentin-en-Yvelines, France
Dieter an Mey, RWTH Aachen University, Germany
Wolfgang E. Nagel, TU Dresden, Germany
Martin Schulz, Lawrence Livermore National Laboratory, USA
Sameer Shende, University of Oregon, USA
Brian Wylie, Jülich Supercomputing Centre, Germany
