Recent work [2] [5] [11] [12] [14] has illustrated the promise of multilevel approaches for partitioning large circuits. Multilevel partitioning recursively clusters the instance until its size is smaller than a given threshold, then unclusters the instance while applying a partitioning refinement algorithm. Our multilevel partitioner uses a new technique to control the number of levels in the matchingbased clustering phase and also exploits recent innovations in classic iterative partitioning [7] [10]. Our heuristic outperforms numerous existing bipartitioning heuristics, with improvements ranging from 6.9 to 27.9% for 100 runs and 3.0 to 20.6% for just 10 runs (while also using less CPU time).
Charles J. Alpert, Jen-Hsin Huang, Andrew B. Kahng