high performance computing on graphics processing units

Co-located with 27th IEEE International Parallel & Distributed Processing Symposium (IPDPS 2013).

his workshop aims to bring the programming community together to explore and discuss various options to make programming heterogeneous systems less challenging and more interesting. The workshop seeks to explore programming methodologies in the form of directive-based approaches, language extensions, novel tools and techniques to create a portable, scalable and productive programming environment. This workshop provides a forum for the presentation of research on all aspects of heterogeneous systems programming models, compiler optimizations, language extensions, and software tools for such systems.

Areas of interest include but are not limited to the following topics:
* Multicore processors and Heterogeneous systems
* Programming models: thread and task based models, data parallel models, stream programming
* Language extensions for GPU programming/environments:
o C/C++ extensions for GPU programming
o OpenMP extensions for Accelerator
o OpenCL/CUDA
o OpenACC
o OpenHMPP
* Compiler optimizations and tuning Heterogeneous systems
o SIMDization/Vectorization
o Parallelization and locality optimizations
o Reducing synchronization and scheduling overheads on GPU and Multicore
o Tiling, parametric tiling and offloading
* Runtime systems for Heterogeneous systems
* Debuggers, and performance analysis tools for Heterogeneous systems
* Operating systems and virtual shared memory for Heterogeneous systems
* Software tools for discovering parallelism
* Application frameworks, Case studies, design patterns, and domain-specific languages for developing manycore applications