Optimized Design of Modular VLSI Chips Using Genetic Algorithms

Abstract

This article introduces the application of genetic algorithms as a means to developing automated design tools for fabrication of modular VLSI design chips. Existing market realities require that product development be fast and predictable. As a result, design flexibility and automation are becoming increasingly important design features. Organizational schemes and automated tools that help engineers deliver a product faster are now needed more than ever before. Modular designs offer a promising solution, as they allow for easy integration of desired features and reconfiguration with a minimum effort. Modules can be swapped in and out of the system because of the precise interface definitions that a modular design imposes. The development of such modular designs has gone hand in hand with that of automated design techniques. Automated design can relieve engineers from the burden of redesign, but can only be effective when it produces optimized solutions. This article proposes the use of genetic algorithms for the construction of an automated design tool for area and speed optimization of modular VLSI chips. Our motivation for the use of genetic algorithms stems from their ability to perform generalized and extensive searches over spaces. The design of modular associative memories, or content-addressable memories (CAMs), is used as a testbed. Given a library of CAM modules, the desired functionality and a set of speed and area constraints, this optimization technique produces a suitable CAM design. The proposed technique is implemented and its performance measures are determined and analyzed.