The workshop will be held in room Willow Glenn II in the Marriott Hotel next to the San Jose Convention Center.

Performance Engineering: A Must for Petascale and Beyond

Abstract Supercomputers are becoming increasingly complex, expensive, and power hungry; the cost of running a large simulation on a top supercomputer can easily reach several M$'s. Therefore, it behooves users to employ these expensive systems as efficiently as possible. To achieve this goal, performance analysis, modeling and simulation should cease to be a skill practiced by a few experts and become an integral part of the development process of large applications. The talk will discuss activities around the Blue Waters project that are aimed as achieving this goal.

Here is a link to Marc Snir's homepage

Session 2: Performance of System Components

2:30-3:00 Comparison of Lock Thrashing Avoidance Methods and Performance Implications for Lock Design, Yan Cui, Yu Chen, and Yuanchun Shi (Tsinghua University, Being, China)

3:00-3:30 Making a Case for Distributed File Systems at Exascale, Ioan Raicu (Illinois Institute of Technology and Argonne National Lab, USA), Ian T. Foster and Pete Beckman (University of Chicago and Argonne National Lab, USA)

3:30-4:00 Break

Session 3: Analysis of Large-scale Systems

4:00-4:30 Visual Analysis of I/O System Behavior for High End Computing Chris Muelder, Carmen Sigovan, Kwan-Liu Ma (University of California, Davis, USA), Jason Cope, Sam Lang, Kamil Iskra, Pete Beckman, and Robert Ross (Argonne National Lab, USA) ( Best Paper Award )

4:30-5:00 Multi-scale Analysis of Large Distributed Computing Systems, Lucas Mello Schnorr, Arnaud Legrand, and Jean-Marc Vincent (INRIA, Grenoble, France)

5:00-5:30 An Analysis of Social Gaming Networks in Online and Face-to-Face Bridge Communities Mihaela Balint, Vlad Posea, Alexandru Dimitriu (Politehnica University of Bucharest, Romania) and Alexandru Iosup, Delft University of Technology, the Netherlands