IWOMP 2010 Final Program [Monday, June 14] 13:00-13:30 Registration 13:30-18:00 Tutorial 18:30- Reception at Restaurant Espoir [Tuesday, June 15] 9:00-9:30 Registration 9:30-9:45 Opening remarks 9:45-10:35 Keynote (Chair: Mitsuhisa Sato) Building the Next Generation of Parallel Applications Michael A Heroux (Sandia National Lab.) For decades parallel computing has been the focus of intense research and development in selected fields, and numerous large-scale parallel applications have been developed. SPMD via MPI has been a dominant approach to parallelism to date, but this approach alone will be insufficient going forward. Presently we are on the threshold of mass deployment of parallelism across most application areas, but the path to developing these applications is uncertain. There are many programming models, languages and architectures from which to pick, and the number of choices is growing. In this presentation we discuss some of the principles of parallel application development that have produced today's codes, and how we can address these principles going forward. We also discuss what must change in order to move forward and give ideas for developing parallel applications now that will have sustained value in the future. 10:35-11:00 --- Coffee Break --- 11:00-12:30 Session 1: Runtime and Optimization (Chair: Matthias S. Mueller) Enabling Low-Overhead Hybrid MPI/OpenMP Parallelism with MPC Patrick Carribault, Marc Perache and Herve Jourdren A ROSE-based OpenMP 3.0 Research Compiler Supporting Multiple Runtime Libraries Chunhua Liao, Daniel J. Quinlan, Thomas Panas and Bronis R. de Supinski Binding Nested OpenMP Programs on Hierarchical Memory Architectures Dirk Schmidl, Christian Terboven, Dieter an Mey and Martin Buecker 12:30-14:00 --- Lunch --- 14:00-16:00 Session 2: Scheduling and Performance (Chair: Alistair Rendell) SWOMPS: A System-Wide OpenMP Scheduler for Many-core Processors Jianian Yan, Jiangzhou He, Wentao Han, Wenguang Chen and Weimin Zheng Topology-aware OpenMP Process Scheduling Peter Thoman, Hans Moritsch and Thomas Fahringer How to reconcile event-based performance analysis with tasking in OpenMP Daniel Lorenz, Bernd Mohr, Christian Roessel, Dirk Schmidl and Felix Wolf Fuzzy Application Parallelization Using OpenMP Chantana Chantrapornchai (Phongpensri), J. Pipatpaisan 16:00-16:30 --- Coffee Break --- 16:30-18:00 Poster indexing & Poster session Towards A Quasi High Level Compiler Comparative and Attributive Model for OpenMP Programs Mohammed Mokbel, Robert Kent, and Michael Wong Flexible Fine Grain Threads Management By StackThreads/Mp Library for OpenMP Task Implementation Adnan and Mitsuhisa Sato A Stream-Computing Extension to OpenMP Antoniu Pop and Albert Cohen Current and Future OpenMP Application Benchmarks by SPEC HPG SPEC High-Performance Group 18:30- Banquet at Sansui-Tei [Wednesday, June 16] 9:00-9:50 Invited talk 1 (Chair: Mitsuhisa Sato) Accelerating to the Promised Land Michael Wolfe (The Portland Group, Inc.) The OpenMP API provides a portable model for efficient, high level thread-parallel programming across platforms, vendors, operating systems. We want a model with the same advantages to address compute accelerators. In this talk, we explore today's accelerator landscape, along with the perils of current programming methods. We demonstrate why OpenCL, while impressive and important, doesn't already solve the programming problem. We close with a summary of the PGI Accelerator Model, and discuss some of the implementation issues into which we have run. 9:50-10:10 --- Coffee Break --- 10:10-11:40 Session 3: Proposed Extensions to OpenMP (Chair: Kohichiro Hotta) A Proposal for User-defined Reductions in OpenMP Alejandro Duran, Roger Ferrer, Michael Klemm, Bronis R. de Supinski and Eduard Ayguade An extension to improve OpenMP tasking control Eduard Ayguade, James Beyer, Alejandro Duran, Roger Ferrer, Grant Haab, Kelvin Li and Federico Massaioli Towards an Error Model for OpenMP Michael Wong, Michael Klemm, Alejandro Duran, Tim Mattson, Grant Haab, Bronis R. de Supinski and Andrey Churbanov 11:40-13:00 --- Lunch --- 13:00-13:50 Invited talk 2 (Chair: Mitsuhisa Sato) Memory Models and OpenMP Hans Boehm (HP Labs) Although multithreaded programming languages are common, there has been a surprising amount of confusion surrounding the basic meaning of shared variables. This has spanned various programming languages and shared memory parallel programming APIs. There is finally a growing consensus, both that programming languages should by default guarantee an interleaving-based semantics (sequential consistency) for data-race-free programs, and on what that should mean. We discuss the motivation for, and consequences of, this guarantee. We briefly survey where various programming languages stand with respect to this guarantee. The memory model description used in OpenMP 3.0 is largely compatible with such an approach, in spite of the very different terminology used in the specification. We argue however that the current reliance on explicit and implicit "flush" operations has some unfortunate consequences. In particular, the current atomics directives are both relatively difficult to use and their mere existence may have unexpected performance consequences, even if they're not used at all. We propose that future OpenMP specifications follow a somewhat different model, more similar to that used in the current draft C and C++ standards. 13:50-14:10 --- Coffee Break --- 14:10-15:40 Session 4: Hybrid programming and Accelerators with OpenMP (Chair: Lawrence Meadows) Hybrid Parallel Programming on SMP Clusters using XPFortran Yuanyuan Zhang, Hidetoshi Iwashita, Kuninori Ishii, Masanori Kaneko, Tomotake Nakamura and Kohichiro Hotta A Case for Including Transactions in OpenMP Michael Wong, Barna Bihari, Bronis R. de Supinski, Peng Wu, Maged Michael, Yan Liu and Wang Chen OMPCUDA : OpenMP Execution Framework for CUDA Based on Omni OpenMP Compiler Satoshi Ohshima, Shoichi Hirasawa and Hiroki Honda 15:40-16:00 --- Coffee break --- 16:00-17:30 Panel session Status of OpenMP 3.1 and 4.0 Specifications Bronis R. de Supinski Can OpenMP Play a Role at Exascale? Moderator: Barbara M. Chapman Panelists: Taisuke Boku, Kohichiro Hotta, Matthias S. Mueller, Michael Wolfe 17:30- Closing