1. An Efficiently Hardware-Software Partitioning for Embedded Multiprocessor FPGA Systems.
- Author
-
Trong-Yen Lee, Yang-Hsin Fan, Yu-Min Cheng, Chia-Chun Tsai, and Rong-Shue Hsiao
- Subjects
FIELD programmable gate arrays ,MULTIPROCESSORS ,GENETIC algorithms ,JPEG (Image coding standard) ,COMPUTER input-output equipment ,COMPUTER software - Abstract
Hardware-software partitioning is a critical issue while developing embedded multiprocessor FPGA systems. In this paper, we propose an efficiently hardware-software partitioning method, namely GHO, that profits the advantages of genetic algorithm and hardware oriented partition to solve the partitioning issues for embedded multiprocessor FPGA systems. In addition, GHO can gain lead to a better partitioning result, faster execution time and lower memory used rate, under satisfied system constraints. Finally, a JPEG encoding system example is used to demonstrate the feasibility for our proposed GHO. Experiment results show that the execution time and memory size are improved by GHO in Xilinx ML310 FPGA platform. [ABSTRACT FROM AUTHOR]
- Published
- 2007