基于约束紧迫度和反向调节算法的硬/软件划分

Hardware/Software Partitioning Based on Constraints-Criticality and Back-Adjusting Algorithm

硬 /软件协同设计的关键问题之一是硬 /软件划分 .本文提出了基于约束紧迫度和反向调节算法的硬 /软件划分方法 .在约束紧迫度算法中 ,首先根据时间紧迫度来决定节点是映射到硬件还是映射到软件 ,然后根据硬 /软件面积的紧迫度选择节点的执行时间 /面积 .反向调节算法则是通过节点之间存在的松弛时间来对系统进行进一步的优化 .实验表明 ,该算法在总体性能上要优于文献〔18〕中提出的 MIBS算法 .

One of the key problems in hardware/software co-design is hardware/software partitioning. The paper presents hardware/software partitioning approach based on constraints-criticality and back-adjusting algorithm. In constraint-criticality algorithm, first the nodes are mapped to hardware or software according to time criticality, then times/areas of the nodes are selected by hardware/software area criticality. Back-Adjusting algorithm makes use of slack time of among nodes to optimize system design still further. Experiment shows that the algorithm is usually better than MIBS presented by paper 〔18〕.