Organizers

• Rusty Lusk (Argonne National Lab), "lusk" AT "mcs.anl.gov"
• Pete Beckman (Argonne National Lab), "beckman" AT "mcs.anl.gov"

Abstract

• Become familiar with the architecture, operation, and usability issues for each of the DOE Leadership Class Facilities.
• Understand application scaling bottlenecks on the systems.
• Learn strategies for achieving good performance with message passing and I/O libraries.
• Explore new programming models, languages, and techniques that can provide scalable performance.
• Learn the tools and suggest strategies for understanding the performance of petascale applications.

Tentative Agenda

Day 1 - Monday, July 27

• 7:45 - 8:30: Breakfast
• 8:30 - 10:00: Users present 10 minutes each about their projects
  • Parallel Performance for Reacting Flow Using, Andy Nonaka (Center for Computational Sciences and Engineering, Lawrence Berkeley National Laboratory)
  • High Resolution Climate Modeling with Geodesic Grids, Don Dazlich, David Randall (Colorado State University)
  • Extended MHD plasma simulation for fusion, Linda Sugiyama (MIT)
  • Madness, Rick Archibald (Computer Science and Mathematics Division, ORNL)
  • Scaling to 150K cores: recent algorithm and performance engineering developments enabling XGC1 to run at scale, M.F. Adams, Columbia University; S. Ku, J. Cummings, P. Worley, E. D’Azevedo, C.S. Chang, and the CPES team (SciDAC Center for Plasma Edge Simulation)
  • FronTier Challenges toward Petascale Computing, Xiaolin Li and the FronTier team (J. Glimm, X. Jiao, R. Samulyak, W. Bo, B. Fix, R. Kaufman, et al) Department of Applied Math and Statistics SUNY at Stony Brook
  • An overview of the R-matrix electron-impact scattering package at NERSC, Connor Balance (Auburn University)
  • Combining QC Codes, Alexander Gaenko (Iowa State University)
  • Petascale Analysis and Visualization, Dave Pugmire (Oak Ridge National Laboratory)
  • Lattice QCD on Leadership-class Machines, M.B. Oktay (MILC Collaboration)
• 10:00 - 10:30: Break
• 10:30 - 12:00: Users present 10 minutes each about their projects
  • High Performance Systems Biology Tools, Peter Graf, Christopher Chang, David Alber, Monte Lunacek, Ambarish Nag, Kwiseon Kim, Michael Siebert (National Renewable Energy Laboratory); Dave Biagioni, David Bortz (University of Colorado, Boulder); Glenn Murray (Colorado School of Mines)
  • Performance bottlenecks in FLASH, Chris Daley (The Center for Astrophysical Thermonuclear Flashes)
  • Granlibakken HPC workshop 2009, Gary Wesenberg (University of Wisconsin‐Madison, BACTER Institute)
  • Common Component Architecture, Kosta Damevski (Department of Mathematics and Computer Science, Virginia State University)
  • Multiphase flow simulation with MFIX, Jeff Dietiker (National Energy Technology Laboratory)
  • Massively Parallel simulation of combustion in gas turbines, Isabelle Dast (European Centre for Research and Advanced Training in Scientific Computation)
  • SNe Ia Simulations with CASTRO, Shawfeng Dong (UCO/Lick Observatory, UCSC)
  • Parallel Computing for Nucleon-Nucleus Scattering, Ian Thompson (Lawrence Livermore National Laboratory)
  • S3D Performance Priorities, Ed Richardson (Sandia National Labs)

• 12:00 - 1:30: Lunch
• 1:30 - 2:15: Overview of ORNL Cray XT, performance, I/O, usage, Vinod Tipparaju
• 2:15 - 3:00: NERSC Overview CSADS Workshop on PetaScale Applications and Performance Strategies, Katie Antypas
• 3:00 - 3:30: Break
• 3:30 - 4:00: Overview of the Argonne Leadership Computing Facility, Scott Parker
• 4:00 - 4:30: Discuss what users want most to cover, their concerns, etc.

• 6:30-8:00: Dinner

Day 2 - Tuesday, July 28

• 7:45 - 8:30: Breakfast
• 8:30 - 10:00: Programming in MPI for Performance and Programming Models for HPC (part 1), Rusty Lusk
• 10:00 - 10:30: Break
• 10:30 - 12:00: Programming in MPI for performance (part 2), Rusty Lusk

• 12:00 - 1:30: Lunch
• 1:30 - 3:00: OpenMP Tutorial(part 1), Lei Huang (University of Houston
• 3:00 - 3:30: Break
• 3:30 - 4:00: OpenMP (part 2), Lei Huang
• 4:00 - 5:00: Parallel Performance Evaluation With TAU, Wyatt Spear (University of Oregon)

• 5:00-7:00: Dinner
• 7:00 - ???: Late Night hacking: study codes and challenges.
  

Day 3 - Wednesday, July 29

• 7:45 - 8:30: Breakfast
• 8:30 - 9:30: HPCToolkit: Sampling-based performance tools for leadership computing, John Mellor-Crummey (Rice)
• 9:30 - 10:00 Coarray Fortran: Past, present and future (part 1) John Mellor-Crummey (Rice)
• 10:00 - 10:30 Break
• 10:30 - 11:00 Coarray Fortran: Past, present and future (part 2)
• 11:00 - 12:00 Parallel I/O in Practice (part I), Rob Ross (Argonne National Laboratory)

• 12:00 - 1:30: Lunch
• 1:30 - 7:00: Hands on work with MPI, performance tools, MPI-IO, ...

• 7:00 - 8:30: Dinner
• 8:30- ???: Late night hacking: continue collaborative work

Day 4 - Thursday, July 30

• 7:45 - 8:30: Breakfast
• 8:30 - 9:30: Introduction to UPC, Rajesh Nishtala
• 9:30 - 10:00: Parallel I/O (part 2), Rob Ross
• 10:00 - 10:30: Break
• 10:30 - 11:00 Parallel I/O (part 3), Rob Ross
• 11:00 - 12:30: Visualization, Mark Hereld (Argonne National Laboratory)

• 12:30-2:00: Lunch
• 2:00-5:30: Hands on work with MPI, performance tools, MPI-IO, ...
• 5:30-6:00: Final discussion and workshop wrap-up

• Dinner
• 8:30- ???: Late night hacking: continue collaborative work

Sponsors