The great commercial success of Graphics Processing Units (GPUs) coupled with increasing programmability has led to a wider range of applications for these chips, well beyond graphics processing. Using GPUs for general purpose processing is now coined as GP-GPU, a field of study and research that is attracting a great deal of interest worldwide.
The aim of this theme is to develop compilers which parse domain-specific high level notations into efficient GPU C code, as well as explore the processing power of GPUs in general purpose computing through the implementation of a number of algorithms in biological sequence analysis, computational biology, financial computing and other applications on NVIDIA GPUs. The results of these implementations will be benchmarked against other hardware e.g. FPGA implementations of the same algorithms, as well as software implementations on standard desktop computers.
Selected Publications
- Cheng Ling, Khaled Benkrid and Tsuyoshi Hamada, ‘A Parameterisable and Scalable Smith-Waterman Algorithm Implementation on CUDA-compatible GPUs’, Proceedings of the 7th IEEE Symposium on Application Specific Processors, San Francisco, CA, USA, July 27th - 28th 2009
- T. Hamada, K. Nitadori, K. Benkrid, Y. Ohno, G. Morimoto, Y. Shibataa, K. Oguria, and M. Taiji, 'A Novel Multiple-Walk Parallel Algorithm for the Barnes-Hut Treecode on GPUs – Towards Cost Effective, High Performance N-Body Simulation', Proceedings of the International Supercomputing Conference, Hamburg, Germany, June 23-26, 2009
Supervisor: Dr. Khaled Benkrid
Collaborator: Dr. Tsuyoshi Hamada, Nagasaki University