• 2013 ESR PI Meeting
    10/09/2013 - 10/11/2013
    Washington
    District of Columbia

2013 ESR PI Meeting

The DOE Office of Advanced Scientific Computing Research (ASCR) and NNSA Advanced Simulation and Computing (ASC) program are sponsoring the Extreme-Scale Research Principal Investigators (PI) Meeting on October 9-11 to help foster the unprecedented level of collaboration that is necessary for extreme-scale computing to realize its potential. Research teams currently funded through one of the ASCR or ASC extreme-scale research programs will join other invited stakeholders to coordinate efforts. Space for this event is strictly limited.

Goals & Expected Outcomes:

  1. Participants are updated about current status and plans in DOE extreme-scale research programs.
  2. Participants are exposed to the current ASCR and ASC extreme-scale research portfolio to gain a broader picture of the evolving challenges and requirements for extreme-scale science.
  3. Participants will identify needs, opportunities, and mechanisms for technical coordination in order to enhance collaboration and maximize research and development output.

Venue

Washington Hilton
1919 Connecticut Ave NW
Washington, DC 20009
United States

Agenda

Wednesday, October 9, 2013 (9:30am - 6:00pm)

  • Welcome and Opening Remarks
  • Exascale and Coordination
  • Advanced Simulation and Computing (ASC) Program Updates
  • Advanced Scientific Computing Research (ASCR) Program Updates
  • Plan for Parallel Working Sessions
  • Lunch
  • Toward Improved Application Resilience
  • User Experiences with Faults and Future Wants
  • Parallel Working Sessions #1
    • Application Foundations: Intra-Plexus Coordination
    • Software Stack Integration
    • Workflow and Energy Management
    • Data Management: Intra-Nexus Coordination
  • Poster/Demo Session #1
  • (6:15-7:15 pm) ASCR Computer Science Meeting (required for all researchers with ASCR Computer Science funding))

Thursday, October 10, 2013 (8:30am - 6:00 pm)

  • Plenary Talks on Exascale Math
  • Parallel Working Sessions #2
    • Programming Models, Programming Environments and Performance Execution
    • Operating and Runtime Systems and Hardware Architectures
    • Data Management and Data Analytics
    • Co-Design and Software Stack
  • Lunch
  • Co-Design Updates
  • Parallel Working Sessions #3
    • Co-Design and Hardware Performance Modeling
    • Software Stack and Data Management and Data Analytics
    • Data Path and Resilience
    • Exascale Math and Application Transition
  • Poster/Demo Session #2

Friday, October 11, 2013 (8:30am - 12:30pm)

  • Reports from Parallel Working Sessions
  • Closing Remarks

Resources

Quad Charts

BSM Quad Chart Oct 2013
-Employ highly parallel discrete-event simulation to explore the characteristics of potential exascale hardware and software architectures. -Explore...
Adding Data Management to Parallel File Systems
Data Analysis at Extreme
ExaHDF5 Quad Chart Oct 2013
FOX Quad Chart Oct 2013
GVR Quad Chart Oct 2013
HPVA Quad Chart Oct 2013
InfoTheory Quad Chart Oct 2013
MEMs Quad Chart Oct 2013
PIPER Quad Chart Oct 2013
RSVP Quad Chart Oct 2013
SDMX Quad Chart Oct 2013
SLEEC Quad Chart Oct 2013
Traleika Quad Chart Oct 2013
Xanalytics Quad Chart Oct 2013
XPRESS Quad Chart Oct 2013

Posters

High-performance computing relies on ever finer threading. Recent advances in processor technology include greater numbers of cores, hyperthreading,...
Modern computational and experimental sciences face a major challenge in coping with the sheer volume and complexity of data produced. Storing,...
Goals: • Create portable application-­system partnership for resilience • Explore efficient implementation of resilient and multi-­version data •...
Traleika Poster Oct 2013

Handouts

This project sets the stage for establishing an integrated approach to modeling performance and power, which will take us “beyond the standard”...
High-performance computing relies on ever finer threading. Recent advances in processor technology include greater numbers of cores, hyperthreading,...
Modern computational and experimental sciences face a major challenge in coping with the sheer volume and complexity of data produced. Storing,...
The energy, parallelism, and error rate projections (≈ 2 × 109 FITS/billion hours or 30 minutes MTTI) for exascale systems represent extraordinary...
As scientists eagerly anticipate the benefits of extremescale computing, roadblocks to science discovery at scale threaten to impede their progress....
Evaluation, Optimization, and Application of Execution Models for Exascale Computing Nek-Bone (developed by Argonne National Laboratory and...
Exascale architectures and applications will be far more complex than today's systems. To harness their power, applications, libraries, and system...
The advent of the exascale era is forcing a re-examination of the computational processes for science, and particularly those for scientific data...
As computers continue their exponential growth in computing power, they also produce and consume an ever increasing amount of data. This increase in...
Modern scientific applications are built out of numerous libraries, spanning a variety of domains. Finding the most efficient way to use and compose...
The Traleika Glacier project will research and mature software technologies addressing major Exascale challenges, and get ready to intercept by the...
Goal: The goal of the XPRESS project is to conduct the research and development of OpenX, a complete system software architecture for Exascale...