Computer Science

Exascale Tools Workshop

2011 Exascale Tools Workshop
October 13-14, 2011
Annapolis, MD
 
   ADDITIONAL INFORMATION
Workshop Report.pdf file (1.2MB)
Workshop Challenges »
Workshop Agenda.pdf file (86KB)
Presentations »
Workshop Participants.pdf file (54KB)
 
Panel Questions:
Performance Analysis Tools »
Debugging and Correctness Tools »
Cross-cutting Issues »
WORKSHOP COMMITTEE:
John Mellor-Crummey Barton Miller
Jeff Hollingsworth Dan Quinlan
Martin Schulz Paul Hovland
Jeff Vetter David Skinner
John Daly
 
ASCR POINT OF CONTACT
Sonia R. Sachs (bio)
 
WORKSHOP GOALS:

This workshop will explore the challenges for Exascale tools and identify promising approaches to meeting these challenges and measuring their progress. A goal will be to define objective criteria for assessing tool features, their usability, and their interaction with compilers, runtime systems, and hardware.  

The workshop will identify general challenges and solutions for the different categories of tools, as well as the specific challenges and solutions for each category.  The workshop will also identify cross-cutting solutions, such as power, resilience, and criteria for measuring tools impact on applications.

At the end of this workshop we will have clear articulation and prioritization of challenges (what we know, and what remains uncertain), prioritized list of responses, roadmap with timelines for implementing those responses, and a rough idea about the order of magnitude of costs involved.

A report that summarizes the workshop discussions and provides recommendations on the programming directions will be generated by the workshop committee.  The final report will be delivered by November 15, 2011.

Workshop Goals Summary:

  1. Define objective criteria for assessing tools for Exascale application development (debugging, correctness, and performance) along a wide spectrum of challenges, namely power, locality, resilience, etc.)
  2. Identify tools requirements and interfaces for hardware and system stack, particularly compilers and runtime systems and metrics for success. Identify integrated systems solutions rather than ad-hoc tools solutions
  3. Prioritize challenges for Exascale tools, both generally and specifically to each tool category
  4. Prioritize options addressing the identified challenges
  5. Lay out a roadmap, with options, timeline, and rough cost estimates for needed Exascale research in the different tools category.

 

Last modified: 3/15/2013 5:35:17 PM