In order to view a complete list of the accepted submissions and their abstracts, please click here
| 8:30am - 4:30pm | TUTORIAL More details are available here For location information, please click here |
| 7:00pm - 9:00pm | After Dinner Reception Purdue Memorial Union, Anniversary Drawing Room |
For location information, please click
here
| 8:00am - 8:45am | Breakfast and Registration |
| 8:45am - 9:00am | Opening Remarks |
| 9:00am - 10:00am | Keynote Programming with Transactions Kunle Olukotun Stanford University (Abstract) Transactional Memory (TM) is emerging as a promising technology that can reduce the complexity of parallel programming for shared-memory multiprocessors. With TM, programmers simply declare that code blocks that manipulate shared data should execute as atomic and isolated transactions with respect to all other code. Concurrency control among multiple parallel transactions is the responsibility of the system. Following a quick overview of the benefits of TM, I will describe how TM programming constructs can be embedded in Java and used to extend OpenMP. |
| 10:00am - 10:30am | ------------------------- BREAK ------------------------- |
| 10:30am - 12:10pm | Session 1: OpenMP Overheads, Hybrid Models Session Chair: Larry Meadows, Intel, OpenMP CEO A Microbenchmark Study of OpenMP Overheads Under Nested Parallelism Vassilios Dimakopoulos, Panagiotis Hadjidoukas and Giorgos Philos CLOMP: Accurately Characterizing OpenMP Application Overheads Greg Bronevetsky, John Gyllenhaal and Bronis deSupinski Detection of Violations of the MPI Standard in Hybrid OpenMP/MPI Applications Tobias Hilbrich, Matthias Mueller and Bettina Krammer Early Experiments with the OpenMP/MPI Hybrid Programming Model Ewing Lusk and Anthony Chan |
| 12:10pm - 1:30pm | -------------------------- LUNCH -------------------------- |
| 1:30pm - 3:10pm |
Session 2: OpenMP for Clusters Session Chair: Rusty Lusk, Argonne National Laboratory First Experiences with Intel Cluster OpenMP Christian Terboven, Dieter an Mey, Dirk Schmidl and Marcus Wagner Micro-benchmarks for Cluster OpenMP Implementations: Memory Consistency Costs H'sien Jin Wong, Jie Cai, Alistair Rendell and Peter Strazdins Incorporation of OpenMP Memory Consistency into Conventional DataFlow Analysis Ayon Basumallik STEP: a distributed OpenMP for coarse-grain parallelism tool Daniel Millot, Alain Muller, Christian Parrot and Frederique Silber-Chaussumier |
| 3:10pm - 3:40pm | -------------------------- BREAK ---------------------------- |
| 3:40pm - 5:10pm | Panel Is OpenMP irrelevant next to MPI and Pthreads? Panel Moderator: Barbara Chapman Panelists: Ananth Grama, Purdue University Rusty Lusk, Argonne National Laboratory Tim Mattson, Intel Corp. Kunle Olukotun, Stanford University Michael Wolfe |
| 6:30pm | IWOMP Dinner For location details, please click here |
For location information, please click
here
| 8:00am - 8:45am | Breakfast and Registration |
| 8:45am - 9:30am | Invited talk Taking OpenMP beyond HPC Timothy G. Mattson Intel Corp. (Abstract) OpenMP was born at a time when parallel computing focused exclusively on high performance computing (HPC) applied to science and engineering. As multi-core chips dominate the market and the core counts grow, parallel computing has gone mainstream ... and OpenMP needs to as well. In this paper, we explore the issues central to mainstream, general purpose computing all too often deemphasized by HPC. We then consider OpenMP relative to each of these issues; highlighting where OpenMP needs to evolve to take it beyond HPC. |
| 9:30am - 10:00am | -------------------------- BREAK ---------------------------- |
| 10:00am - 11:40am | Session 3: OpenMP Tasking Models and Extensions Session Chair: Rick Kufrin, NCSA/University of Illinois Evaluation of OpenMP Task scheduling strategies Alejandro Duran, Julita Corbalan and Eduard Ayguade OpenMP Extensions For Generic Libraries Prabhanjan Kambadur, Douglas Gregor and Andrew Lumsdaine Streams: emerging from a shared memory model Benedict Gaster Extending the OpenMP Tasking Model to Allow Dependent Tasks Alex Duran, Josep M. Perez, Eduard Ayguade, Rosa Badia and Jesus Labarta |
| 11:40am - 1:00pm | -------------------------- LUNCH ---------------------------- |
| 1:00pm - 2:40pm |
Session 4: Applications, Scheduling, Tools Session Chair: Ruud Van der Pas, Sun Microsystems, Menlo Park, CA On Multi-Threaded Satisfiability Solving with OpenMP Pascal Vander-Swalmen, Gilles Dequen and Michael Krajecki Parallelism and Scalability in an Image Processing Application Morten S. Rasmussen, Matthias B. Stuart and Sven Karlsson Scheduling Dynamic OpenMP Applications over Multicore Architectures Francois Broquedis, Francois Diakhate, Samuel Thibault, Olivier Aumage, Raymond Namyst and Pierre-Andre Wacrenier Visualizing the Program Execution Control Flow of OpenMP Applications Karl Fuerlinger and Shirley Moore |
| 2:40pm - 3:10pm | -------------------------- BREAK ---------------------------- |
| 3:10pm - 4:40pm | "OpenMP: Status and Future" OpenMP ARB, Vendors Session Chairs: Larry Meadows, Intel Corp. Sanjiv Shah, Intel Corp. |
*This program is tentative and subject to change.
In order to view a complete list of the accepted submissions and their abstracts, please click here