Personal tools
You are here: Home Workshops Summer 2008 Workshop on Leadership-class Machines, Petascale Applications, and Performance Strategies
Document Actions

Workshop on Leadership-class Machines, Petascale Applications, and Performance Strategies

by admin last modified 2008-07-18 12:01

July 14-17, 2008, Snowbird Ski and Summer Resort, Snowbird, Utah, USA

Organizers

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


Abstract


Goals:
  • 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.


Agenda

Day 1 - Monday, July 14

Morning
  • 7:45 - 8:30: Continental Breakfast
  • 8:30 - 10:00: Users present 10-15 minutes each about their projects
  • 10:00 - 10:30: Break
  • 10:30 - 12:00: Users present 10-15 minutes each about their projects
Afternoon
  • 12:00 - 1:30: Lunch on your own
  • 1:30 - 2:15: Vinod Tipparaju: Overview of ORNL machine, performance, I/O, usage,
  • 2:15 - 3:00: Katie Antypas: Overview of NERSC machine, performance, I/O, usage, etc.
  • 3:00 - 3:30: Break
  • 3:30 - 4:00: Pete Beckman: Overview of BG/L, BG/P, what ANL will be getting
  • 4:00 - 4:30: Discuss what users want most to cover, their concerns, etc.
Evening
  • 6:30-8:00 Working buffet dinner: discuss application challenges


Day 2 - Tuesday, July 15

Morning
  • 7:45 - 8:30: Continental Breakfast
  • 8:30 - 10:00: Rusty Lusk: Programming in MPI for performance (part 1)
  • 10:00 - 10:30: Break
  • 10:30 - 12:00: Rusty Lusk: Programming in MPI for performance (part 2)
Afternoon
  • 12:00 - 1:30: Lunch on your own
  • 1:30 - 3:00: Rob Latham: Parallel I/O (part 1)
  • 3:00 - 3:30: Break
  • 3:30 - 4:00: Rob Latham: Parallel I/O (part 2)
  • 4:00 - 5:00: Alan Morris: TAU
Evening
  • 5:00-7:00 Dinner on your own
  • 7:00 - ???: Late Night hacking: study codes and challenges.


Day 3 - Wednesday, July 16

Morning
  • 7:45 - 8:30: Continental Breakfast
  • 8:30 - 10:00: John Mellor-Crummey: HPCToolkit performance tools
  • 10:00 - 10:30: Break
  • 10:30 - 12:00: Oscar Hernandez: OpenMP
Afternoon
  • 12:00 - 1:30 Lunch on your own
  • 1:30 - 7:00 Hands on work with MPI, performance tools, MPI-IO, ...
Evening
  • 7:00 - 8:30 Working buffet dinner: discuss afternoon's activities and challenges
  • 8:30- ??? Late night hacking: continue collaborative work


Day 4 - Thursday, July 17

Morning
  • 7:45 - 8:30: Hot Breakfast
  • 8:30 - 9:30: Rajesh Nishtala: UPC / Titanium
  • 9:30 - 10:00: John Mellor-Crummey: CoArray Fortran (part 1)
  • 10:00 - 10:30: Break
  • 10:30 - 11:00 John Mellor-Crummey: CoArray Fortran (part 2)
  • 11:00 - 12:30: Mark Hereld: Visualization and Data Analysis
Afternoon
  • 12:30-2:00 Lunch on your own
  • 2:00-5:30 Hands on work with MPI, performance tools, MPI-IO, ...
  • 5:30-6:00 Final discussion and workshop wrap-up
Evening
  • Dinner on your own
  • 8:30- ??? Late night hacking: continue collaborative work


Sponsors

This workshop is sponsored by the Center for Scalable Application Development Software, with funding from the DOE Office of Science through the Scientific Discovery through Advanced Computing (SciDAC) program.

« May 2017 »
Su Mo Tu We Th Fr Sa
123456
78910111213
14151617181920
21222324252627
28293031
 

Powered by Plone

CScADS Collaborators include:

Rice University ANL UCB UTK WISC