An efficient eigenvector-node interchange approach for finding netlist partitions

A fast eigenvector technique for obtaining good initial node partitions of netlists for use in interchange heuristics is described. The method is based on approximating the netlist or hypergraph by a weighted graph G and applying the eigenvector technique of E.R. Barnes (1982) to partition G and k blocks of fixed module size. An efficient generalization of the Fiduccia-Mattheyses node-interchange heuristic is developed to further reduce the number of nets connecting k blocks. This node-interchange heuristic is tested on the one resulting netlist partition obtained by this eigenvector approach on a variety of small to large sized benchmark netlist partitioning problems (200 to 12000 modules and nets). The test results show that this eigenvector-node-interchange approach yields netlist partitions that are competitive with the best netlist partitions obtained by using node-interchange heuristics alone on many random initial netlist partitions. The running time of this method is a small fraction of that previous node-interchange methods