A Guiding Function Based Greedy Partitioning Algorithm for Dynamically Reconfigurable Systems

Hardware/software (hw/sw) partitioning is a critical step in the co-design, it directly affects the system performance largely. Targeting the shortcomings existed in the traditional algorithms, in this paper a guiding function based greedy partitioning algorithm (GFBGPA) is presented. The algorithm especially aims at the parameterized FPGA coprocessors architecture. The guiding function is determined by the criticality, speedup, execution probability and area of the task, which not only improves the hardware resource utilization effectively but also takes care of the task on the critical path at the same time, so the algorithm can get an appropriate optimum solution. In the end, this paper analyses the experiment results