TOP
Search the Dagstuhl Website
Looking for information on the websites of the individual seminars? - Then please:
Not found what you are looking for? - Some of our services have separate websites, each with its own search option. Please check the following list:
Schloss Dagstuhl - LZI - Logo
Schloss Dagstuhl Services
Seminars
Within this website:
External resources:
  • DOOR (for registering your stay at Dagstuhl)
  • DOSA (for proposing future Dagstuhl Seminars or Dagstuhl Perspectives Workshops)
Publishing
Within this website:
External resources:
dblp
Within this website:
External resources:
  • the dblp Computer Science Bibliography


Dagstuhl Seminar 07341

Code Instrumentation and Modeling for Parallel Performance Analysis

( Aug 19 – Aug 24, 2007 )


Permalink
Please use the following short url to reference this page: https://www.dagstuhl.de/07341

Organizers
  • Adolfy Hoisie (Pacific Northwest National Lab. - Richland, US)
  • Barton P. Miller (University of Wisconsin - Madison, US)
  • Bernd Mohr (Jülich Supercomputing Centre, DE)



Summary

Given the exponential increase in the complexity of nowadays parallel systems, parallel applications often fail to exploit the full power of the underlying hardware. At scale, it is not uncommon for applications to run at parallel efficiencies in the low single digits. Moreover, their optimization is extremely difficult due to the inherent complexity of the systems and of the applications themselves. Therefore, a variety of projects aim to develop tools for the measurement, analysis, and visualization of parallel program performance in order to help and guide users in the optimization process.

This meeting was the third in a series of seminars related to the topic “Performance Analysis of Parallel and Distributed Programs”, with previous meetings being the Dagstuhl Seminar 02341 on “Performance Analysis and Distributed Computing” held in August 2002 and Seminar 05501 on “Automatic Performance Analysis” in December 2005. While these seminars concentrated on the “analysis” part of performance analysis, at the most recent seminar the focus was on the building blocks of program instrumentation and modeling that are prerequisites for the analysis phase. As a result, the presentations of the participants concentrated on several fundamental issues related to instrumentation for generating highquality performance data, methodologies for performance modeling leading to accurate predictions for the performance, and on the ways in which these techniques are combined for the performance analysis of applications and systems.

The program consisted of 28 presentations and practical tool demonstrations as well as two “open mic” sessions where time was set aside for spontaneous discussions and ”brain storming”. The seminar brought together a total of 48 researchers and developers working in the area of performance from universities, national research laboratories and, especially important, from three major computer vendors. The goals were to increase the exchange of ideas, knowledge transfer, foster a multidisciplinary approach to attacking this very important research problem with direct impact on the way in which we design and utilize parallel systems to achieve high application performance.

The presentations can be grouped thematically as follows:

  • Session “Performance Analysis in General”
  • Session “Instrumentation”
  • Session “Modeling”
  • Session “Scalability”
  • Session “Tools”
  • “Open Mic” Sessions
  • Session “Short Announcements”

Despite the larger than normal number participants, the seminar was very successful due to the dedicated professionalism and discipline of the participants on one side and the very helpful and professional staff of Dagstuhl on the other side. Lively discussions and spontaneous computer demonstrations continued every day well beyond midnight. It is important to note that the group meeting and residential aspects of Dagstuhl and the five-day format provide a continuity of thought and discussion unavailable in other conference, workshop, or meeting settings. At Dagstuhl, we have time for considered (and reconsidered!) dialogs whose impact last well beyond the meeting week.

A half-day excursion including a guided tour of “Burg Eltz” and a wine-tasting and dinner at the lovely down of Beilstein on the Mosel river completed the program.


Participants
  • Kevin Barker (Los Alamos National Lab., US)
  • Daniel Becker (Jülich Supercomputing Centre, DE)
  • Andrew Bernat (University of Wisconsin - Madison, US)
  • Holger Brunst (TU Dresden, DE) [dblp]
  • Barbara M. Chapman (University of Houston, US) [dblp]
  • Luiz DeRose (Cray Inc. - Saint Paul, US)
  • Sudip Dosanjh (Sandia National Labs - Albuquerque, US)
  • Markus Geimer (Jülich Supercomputing Centre, DE) [dblp]
  • Vladimir S. Getov (University of Westminster - London, GB)
  • Judit Gimenez (Barcelona Supercomputing Center, ES) [dblp]
  • Adolfy Hoisie (Pacific Northwest National Lab. - Richland, US) [dblp]
  • Jeffrey K. Hollingsworth (University of Maryland - College Park, US) [dblp]
  • Kevin A Huck (Barcelona Supercomputing Center, ES)
  • Heike Jagode (University of Tennessee, US)
  • Terry Jones (LLNL - Livermore, US)
  • Guido Juckeland (TU Dresden, DE)
  • Matthias Jurenz (TU Dresden, DE)
  • Rainer Keller (Oak Ridge National Laboratory, US)
  • Darren Kerbyson (Pacific Northwest National Lab. - Richland, US)
  • David Klepacki (IBM TJ Watson Research Center, US)
  • Michael Kluge (TU Dresden, DE) [dblp]
  • Dieter Kranzlmüller (LMU München, DE) [dblp]
  • Jesus Labarta (Barcelona Supercomputing Center, ES) [dblp]
  • Ewing Lusk (Argonne National Laboratory, US)
  • Allen D. Malony (University of Oregon - Eugene, US) [dblp]
  • Sally A. McKee (Chalmers UT - Göteborg, SE) [dblp]
  • John Mellor-Crummey (Rice University - Houston, US)
  • Barton P. Miller (University of Wisconsin - Madison, US) [dblp]
  • Hartmut Mix (TU Dresden, DE)
  • Bernd Mohr (Jülich Supercomputing Centre, DE) [dblp]
  • Anna Morajko (Autonomus University of Barcelona, ES)
  • Oleg Morajko (Autonomus University of Barcelona, ES)
  • Frank Mueller (North Carolina State University - Raleigh, US) [dblp]
  • Matthias S. Müller (TU Dresden, DE) [dblp]
  • Heidi Poxon (Cray Inc. - Saint Paul, US) [dblp]
  • Radu Prodan (Universität Innsbruck, AT)
  • Philip Roth (Oak Ridge National Laboratory, US) [dblp]
  • Simone Sbaraglia (IBM TJ Watson Research Center - Yorktown Heights, US)
  • Martin Schulz (LLNL - Livermore, US) [dblp]
  • Erich Strohmaier (Lawrence Berkeley National Laboratory, US)
  • Jim Tomkins (Sandia National Labs - Albuquerque, US)
  • Jesper Larsson Träff (Universität Wien, AT) [dblp]
  • Jeffrey S. Vetter (Oak Ridge National Laboratory, US) [dblp]
  • Harvey Wasserman (Lawrence Berkeley National Laboratory, US)
  • Felix Wolf (GRS for Simulation Sciences - Aachen, DE) [dblp]
  • Patrick Worley (Oak Ridge National Laboratory, US)
  • Brian J. N. Wylie (Jülich Supercomputing Centre, DE)
  • Mary Zosel (LLNL - Livermore, US)

Related Seminars
  • Dagstuhl Seminar 02341: Performance Analysis and Distributed Computing (2002-08-18 - 2002-08-23) (Details)
  • Dagstuhl Seminar 05501: Automatic Performance Analysis (2005-12-12 - 2005-12-16) (Details)
  • Dagstuhl Seminar 10181: Program Development for Extreme-Scale Computing (2010-05-02 - 2010-05-07) (Details)

Classification
  • sw-engineering
  • optimization/scheduling
  • modeling simulation
  • programming languages / compiler

Keywords
  • program instrumentation
  • performance analysis
  • performance modeling
  • parallel and distributed computing