A hierarchical circuit clustering algorithm with stable performance

Clustering is almost essential in improving the performance of iterative pa rtitioning algorithms. In this paper, we present a clustering algorithm bas ed on the following observation: if a group of cells is assigned to the sam e partition in numerous local optimum solutions, it is desirable to merge t he group into a cluster. The proposed algorithm finds such a group of cells from randomly generated local optimum solutions and merges it into a clust er. We implemented a multilevel bipartitioning algorithm (MBP) based on the proposed clustering algorithm. For MCNC benchmark netlists, MBP improves t he total average cut size by 9% and the total best cut size by 3-4%, compar ed with the previous state-of-the-art partitioners.