A hybrid genetic algorithm for VLSI floorplanning

Abstract

Floorplanning is the first stage of the very large scale integrated-circuit (VLSI) physical design process, the resultant quality of this stage is very important for successive design stages. From the computational point of view, VLSI floorplanning is an NP-hard problem. In this paper, a hybrid genetic algorithm (HGA) for a non-slicing and hard-module VLSI floorplanning problem is presented. This HGA uses an effective genetic search method to explore the search space and an efficient local search method to exploit information in the search region. Experimental results on MCNC benchmarks show that the HGA is effective and promising in building block layout application.