A fast neural-network algorithm for VLSI cell placement

Cell placement is an important phase of current VLSI circuit design styles such as standard cell, gate array, and Field Programmable Gate Array (FPGA). Although nondeterministic algorithms such as Simulated Annealing (SA) were successful in solving this problem, they are known to be slow. In this paper, a neural network algorithm is proposed that produces solutions as good as SA in substantially less time. This algorithm is based on Mean Field Annealing (MFA) technique, which was successfully applied to various combinatorial optimization problems. A MFA formulation for the cell placement problem is derived which can easily be applied to all VLSI design styles. To demonstrate that the proposed algorithm is applicable in practice, a detailed formulation for the FPGA design style is derived, and the layouts of several benchmark circuits are generated. The performance of the proposed cell placement algorithm is evaluated in comparison with commercial automated circuit design software Xilinx Automatic Place and Route (APR) which uses SA technique. Performance evaluation is conducted using ACM/SIGDA Design Automation benchmark circuits. Experimental results indicate that the proposed MFA algorithm produces comparable results with APR. However, MFA is almost 20 times faster than APR on the average.

[1]  Brian W. Kernighan,et al.  A Procedure for Placement of Standard-Cell VLSI Circuits , 1985, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[2]  Tevfik Bultan,et al.  Circuit partitioning using mean field annealing , 1995, Neurocomputing.

[3]  Mattias Ohlsson,et al.  A Study of the Mean Field Approach to Knapsack Problems , 1997, Neural Networks.

[4]  Thomas Lengauer,et al.  Combinatorial algorithms for integrated circuit layout , 1990, Applicable theory in computer science.

[5]  Yoshikane Takahashi,et al.  Mathematical improvement of the Hopfield model for TSP feasible solutions by synapse dynamical systems , 1997, Neurocomputing.

[6]  Carsten Peterson,et al.  Complex Scheduling with Potts Neural Networks , 1992, Neural Computation.

[7]  Naveed A. Sherwani,et al.  Algorithms for VLSI Physical Design Automation , 1999, Springer US.

[8]  Laurent Hérault,et al.  Neural Networks and Graph K-Partitioning , 1989, Complex Syst..

[9]  Tevfik Bultan,et al.  A New Mapping Heuristic Based on Mean Field Annealing , 1992, J. Parallel Distributed Comput..

[10]  Pinaki Mazumder,et al.  A Neural Network Design for Circuit Partitioning , 1989, 26th ACM/IEEE Design Automation Conference.

[11]  Carsten Peterson,et al.  A Potts Neuron Approach to Communication Routing , 1997, Neural Computation.

[12]  A. El Gamal,et al.  Architecture of field-programmable gate arrays , 1993, Proc. IEEE.

[13]  Martin Lagerholm,et al.  A Potts Neuron Approach toCommunication , 1998 .

[14]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[15]  Zvonko G. Vranesic,et al.  Field-Programmable Gate Arrays , 1992 .

[16]  Carsten Peterson,et al.  A New Method for Mapping Optimization Problems Onto Neural Networks , 1989, Int. J. Neural Syst..

[17]  Carsten Peterson,et al.  Neural Networks for Optimization Problems with Inequality Constraints: The Knapsack Problem , 1993, Neural Computation.

[18]  Carl Sechen,et al.  VLSI Placement and Global Routing Using Simulated Annealing , 1988 .

[19]  David E. van den Bout,et al.  Graph partitioning using annealed neural networks , 1990, International 1989 Joint Conference on Neural Networks.

[20]  Pinaki Mazumder,et al.  VLSI cell placement techniques , 1991, CSUR.

[21]  Robert J. Cimikowski,et al.  A neural-network algorithm for a graph layout problem , 1996, IEEE Trans. Neural Networks.

[22]  OhlssonMattais,et al.  Neural networks for optimization problems with inequality constraints , 1993 .