GPUs are cost-effective platforms for computational intensive applications providing significant peak performance. However, it is a major challenge to deliver the intrinsic performance of such architectures to end applications.
This special issue addresses programming approaches (stream programming) and key techniques to leverage the computing power of GPUs. Experiences gained while adapting scientific codes to run efficiently on such architectures are welcome. We solicit unpublished papers about research challenges and advances addressing porting of scientific codes and algorithms to GPU platforms.
Suitable topics include but are not limited to:
- Scientific applications optimised for GPUs
- Programming approaches for multicore CPUs and GPUs
- Stream computing
- Programming environments for GPUs
- Benchmarking/measurements for GPUs
- GPU architectures
- CUDA and OpenCL
Submission deadline: 18 April 2011
Acceptance notification: 20 June 2011
Final paper due: 31 August 2011