Virtual Institute – High Productivity Supercomputing

Workshop on Extreme-Scale Programming Tools

SC13

Date

November 18, 2013

Location

in conjunction with
Supercomputing 2013
Denver, Colorado, USA

As we approach exascale, architectural complexity and severe resource limitations with respect to power, memory and I/O, make tools support in debugging and performance optimization more critical than ever before. However, the challenges mentioned above also apply to the tools development and, in particular, raise importance of topics such as automatic tuning and methodologies for exascale tools-aided application development. This workshop will serve as a forum for application, system, and tool developers to discuss the requirements for future exascale-enabled tools and the roadblocks that need to be addressed on the way. We also highly encourage application developers to share their experiences with using the tools.

The workshop is organized by the Virtual Institute - High Productivity Supercomputing in collaboration with the Priority Programme "Software for Exascale Computing" (SPPEXA) of the German Research Foundation (DFG). The event will also focus on the community-building process necessary to create an integrated tools-suite ready for an exascale software stack.

Workshop topics

  • Programming Tools, for instance, Performance Analysis and Tuning Tools, Debuggers, IDEs
  • Methodologies for Performance Engineering
  • Tool Technologies Tackling Extreme-Scale Challenges, such as Scalability, Resilience, etc.
  • Tool Infrastructures
  • Application Developer Experiences with Programming Tools

Registration

All attendees of the workshop are required to register for the SC Workshop Program on Monday, November 18th. Please register through the official registration at Supercomputing '13.

Agenda

09:00 - 09:20 Welcome and introduction to VI-HPS [PDF]
Michael Gerndt, TU Munich
09:20 - 10:00 Keynote Talk: Tools for Earth System Modelling [PDF]
Thomas Ludwig, German Climate Computing Centre
10:00 - 10:30 Coffee break
10:30 - 10:50 FEPA - A framework for systematic energy and performance analysis of extreme-scale applications in HPC computing centers [PDF]
Jan Treibig, Regional Computing Center Erlangen
10:50 - 11:10 Composing an autotuning toolkit for exascale [PDF]
Manu Shantharam, University of Utah
11:10 - 11:30 ELASTIC: Dynamic Tuning for Large-Scale Parallel Applications [PDF]
Antonio Espinosa, Universitat Autònoma de Barcelona
11:30 - 11:50 Dynamic Tuning of the Energy Consumption [PDF]
David Brayford, Leibniz Supercomputing Centre
11:50 - 12:10 A Load-Balancing Simulator for Software Re-Engineering [PDF]
Monika Lücke, German Research School for Simulation Sciences
12:10 - 12:30 Domain Decomposition and Halo Exchange at 640k MPI Processes [PDF]
Andreas Schäfer, Friedrich-Alexander-Universität Erlangen-Nürnberg
12:30 - 12:40 Discussion
12:40 - 14:00 Lunch
14:00 - 14:40 Keynote Talk: Addressing Performance and Programmability Challenges in Current and Future Supercomputers [PDF]
Luiz De Rose, Cray
14:40 - 15:00 Introducing PAThWay: Structured and methodical performance engineering [PDF]
Isaias Compres, Technical University Munich
15:00 - 15:30 Coffee break
15:30 - 15:50 Engineering Performance Tools for Extreme Scale
Allen D. Malony, University of Oregon
15:50 - 16:10 Two Roads to Extreme Scale Event Trace Analysis? [PDF]
Andreas Knuepfer, ZIH, TU Dresden
16:10 - 16:30 Performance Profiling and Debugging at the Extreme Scale and Beyond [PDF]
David Lecomber, Allinea Software
16:30 - 16:50 Analyzing Future Exascale Platforms on Today’s Machines
Martin Schulz, Lawrence Livermore National Laboratory
16:50 - 17:10 Building on Lessons Learned From Over a Decade of MRNet Research and Development [PDF]
Barton P. Miller, University of Wisconsin-Madison
17:10 - 17:30 Discussion

Time and location

09:00 a.m. to 5:30 p.m.

Conference Room 501,
Colorado Convention Center
Denver, Colorado, USA

Please check the SC'13 workshop page for up-to-date information on the location.

Organizers

VI-HPS SPPEXA

Organizing committee

Michael Gerndt, TU Munich, Germany
Bettina Krammer, MoRitS, Bielefeld University & University of Applied Science Bielefeld, Germany
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
Felix Wolf, German Research School for Simulation Sciences, Germany
Brian Wylie, Jülich Supercomputing Centre, Germany

Contact person

Yury Oleynik (oleynik@in.tum.de)