Talk 5.1: 1:30PM–2:45PM
| Session Name: | Computer Architectures, Compilers and Programming Models |
| Session Time: | 1:30PM–2:45PM |
| Author Name: | Andrew W. H. House |
| Author Email: | ahouse@eecg.toronto.edu |
| Talk Title: | Programming Models and Tools for Multi-FPGA Computing |
| Slides: | 5-1.ppt |
| Abstract: | Modern FPGA technology is being investigated for high performance computing applications due to the large speedup FPGAs can offer over conventional microprocessors. However, in order to surpass the performance of large multiprocessor systems, it is necessary to develop multi-FPGA computing platforms. One of the challenges faced in the development of these new computing platforms is that traditional hardware design flows do not scale well to the dozens or hundreds of FPGAs that will form the basis of such a system. The purpose of this research is to investigate programming models and tools appropriate for these new classes of computing hardware, specifically: the development of a high-level language for synthesis and parallel programming; the development of algorithms for the partitioning of systems across multiple FPGAs; and integration with existing CAD tools to provide an end-to-end solution. |
| Research Group: | Computer |
| Degree Program: | Ph.D. |
| Author Bio: | Andrew House received the B.Eng and M.Eng degrees from Memorial University of Newfoundland, Canada, in 2002 and 2004, respectively. During his M.Eng degree, he completed the Graduate Program in Teaching, taught a course as sessional instructor, and was named a Fellow of the School of Graduate Studies. Since 2004 he has been in the Ph.D program at the University of Toronto, ON, Canada, where his research is in the areas of high performance computing, synthesis of high-level languages, and digital hardware design. |