You are here
Workshop program
Workshop Program
Day 1, Tuesday 12 - Horti Sallustiani
| 9.00 | Registration |
| 9.30 | Opening Remarks |
| 9.45 | Keynote: C++11 Style - A Touch of Class Bjarne Stroustrup, Professor and holder of the College of Engineering Chair in Computer Science, Texas A&M University |
| We know how to write bad code: Litter our programs with casts, macros, pointers, naked new and deletes, and complicated control structures. Alternatively (or in addition), obscure every design decision in a mess of deeply nested abstractions using the latest object-oriented programming and generic programming tricks. For good measure, complicate our algorithms with interesting special cases. Such code is incomprehensible, unmaintainable, usually inefficient, and not uncommon. But how do we write good code? What principles, techniques, and idioms can we exploit to make it easier to produce quality code? I will make an argument for type-rich interfaces, compact data structures, integrated resource management and error handling, and highly-structured algorithmic code. I will illustrate my ideas and motivate my guidelines with a few idiomatic code examples. I will not discuss concurrent systems in any detail, but I will articulate ideas for their construction. I will use C++11 freely. Examples include auto, general constant expressions, uniform initialization, type aliases, type safe threading, and user-defined literals. Developing a "modern style" is essential if we don't want to maintain newly-written 1970s and 1980s style code in 2020. This presentation reflects my thoughts on what "Modern C++" should mean in the 2010s: a language for programming based on light-weight abstraction with a direct and efficient mapping to hardware, suitable for infrastructure code. |
|
| 10.45 | Coffee Break |
| 11.15-12.45 | Session 1: Proposed Extensions to OpenMP |
| 11.15 | Specification and Performance Evaluation of Parallel I/O Interfaces for OpenMP Kshitij Mehta, Edgar Gabriel and Barbara Chapman |
| 11.45 | The Design of OpenMP Thread Affinity Alexandre E. Eichenberger, Christian Terboven, Michael Wong and Dieter an Mey |
| 12.15 | Auto-scoping for OpenMP tasks Sara Royuela, Alejandro Duran, Chunhua Liao and Daniel J. Quinlan |
| 12.45 | Lunch break |
| 14.15-15.45 | Session 2: Proposed Extensions to OpenMP |
| 14.15 | A Case for Including Transactions in OpenMP II: Hardware Transactional Memory Barna L. Bihari, Michael Wong, Amy Wang, Bronis R. de Supinski and Wang Chen |
| 14.45 | Extending OpenMP* with Vector Constructs for Modern Multicore SIMD Architectures Michael Klemm, Alejandro Duran, Xinmin Tian, Hideki Saito, Diego Caballero and Xavier Martorell |
| 15.15 | Coffee break |
| 15.45-16.45 | Session 3: Runtime Environments |
| 15.45 | Automatic OpenMP Loop Scheduling: A Combined Compiler and Runtime Approach Peter Thoman, Herbert Jordan, Simone Pellegrini and Thomas Fahringer |
| 16.15 | libKOMP, an Efficient OpenMP Runtime System for Both Fork-Join and Data-Flow Paradigms François Broquedis, Thierry Gautier and Vincent Danjean |
| 16.45-17.45 | Poster Session |
| 19.30 | Aperitif and Social Dinner |
Day 2, Wednesday 13 - Horti Sallustiani
| 9.15 | Keynote. GPU Pragmas: Enabling the next HPC revolution Ian Buck, General Manager of GPU Computing, NVIDIA. |
| This keynote will cover the evolution of GPU computing, past, present and future and the opportunity that directive based programming is enabling in the HPC and broader markets. Ian will provide a history of GPU Computing, discuss the opportunities modern GPUs represent for compiler developers, and why directives like OpenMP are key to wide spread developer adoption. | |
| 10.15 | Coffee Break |
| 10.45-12.45 | Session 4: Optimization and Accelerators |
| 10.45 | Experiments with WRF on Intel Many Integrated Core (Intel MIC) Architecture Larry Meadows |
| 11.15 | Optimizing the Advanced Accelerator Simulation Framework Synergia Using OpenMP Hongzhang Shan, Erich Strohmaier, James Amundson and Eric G. Stern |
| 11.45 | Using Compiler Directives for Accelerating CFD Applications on GPUs Haoqiang Jin, Mark Kellogg and Piyush Mehrotra |
| 12.15 | Effects of Compiler Optimizations in OpenMP to CUDA Translation Amit Sabne, Putt Sakdhnagool and Rudolf Eigenmann |
| 12.45 | Lunch break |
| 14.15-15.45 | Session 5: Task Parallelism |
| 14.15 | Assessing OpenMP Tasking Implementations on NUMA Architectures Christian Terboven, Dirk Schmidl, Tim Cramer and Dieter an Mey |
| 14.45 | Performance Analysis Techniques for Task-Based OpenMP Applications Dirk Schmidl, Peter Phillippen, Daniel Lorenz, Christian Rössel, Markus Geimer, Dieter an Mey, Bernd Mohr and Felix Wolf |
| 15.15 | Task-based Execution of Nested OpenMP Loops Spiros N. Agathos, Panagiotis E. Hadjidoukas and Vassilios V. Dimakopoulos |
| 15.45 | Coffee break |
| 16.15-17.15 | Session 6: Validation and Benchmarks |
| 16.15 | SPEC OMP2012 - An application benchmark suite for parallel systems using OpenMP Matthias S. Müller, John Baron, William C. Brantley, Huiyu Feng, Daniel Hackenberg, Robert Henschel, Gabriele Jost, Daniel Molka, Chris Parrot, Joe Robichaux, Pavel Shelepugin, Matthijs van Waveren, Brian Whitney and Kalyan Kumaran |
| 16.45 | An OpenMP 3.1 Validation Testsuite Cheng Wang, Sunita Chandrasekaran and Barbara Chapman |
| 17.15 | OpenMP Language Committee Report LC1, Bronis R. de Supinski, Lawrence Livermore National Laboratory LC2, Michael Wong, IBM Software Group |
| 17.45 | Closing Remarks |
