| Paper Abstract and Keywords |
| Presentation |
2006-01-18 15:20
A Performance Improvement Strategy for Numerical Integration on an FPGA-Based Biochemical Simulator ReCSiP
Yuri Nishikawa, Yasunori Osana, Masato Yoshimi, Yow Iwaoka, Toshinori Kojima (Keio Univ.), Akira Funahashi, Noriko Hiroi (JST), Yuichiro Shibata, Naoki Iwanaga (Nagasaki Univ.), Hiroaki Kitano (JST), Hideharu Amano (Keio Univ.) |
| Abstract |
(in Japanese) |
(See Japanese page) |
| (in English) |
Computational simulation is a highly effective solution for cellular analyses in recent bioinformatics. As the scale of analysis objects are rapidly expanding, however, it is a prime task to develop a simulator with high throughput. ReCSiP, an FPGA-based biochemical simulator, is being exploited for this necessity. Further speed-up of the simulator is enhanced with an improvement of its pipelining efficiency, by a parallel operation of a rate-law solver module and a numerical integration module.
This research report discusses the methodology of enhancing the performance of ReCSiP, its implementation and the evaluation result. Speed-up of 1.96x in maximum is achieved with an area increase of 29\% compared to the conventional implementation. |
| Keyword |
(in Japanese) |
(See Japanese page) |
| (in English) |
Biochemical simulator / numerical integration / Euler's method / Pipelining efficiency / FPGA / / / |
| Reference Info. |
IEICE Tech. Rep., vol. 105, no. 518, RECONF2005-95, pp. 53-58, Jan. 2006. |
| Paper # |
RECONF2005-95 |
| Date of Issue |
2006-01-11 (VLD, CPSY, RECONF) |
| ISSN |
Print edition: ISSN 0913-5685 |
| Download PDF |
|