EECE 583 Project - Mapping, Placing, and Routing of MCNC Benchmark Circuits to Quantum Annealer Architecture

Su, J., Tu, T., & He, L. (n.d.). A Quantum Annealing Approach for Boolean Satisfiability Problem. https://doi.org/10.1145/2897937.2897973

Presentation available here

This repo includes the source code for a mapping, placing, and routing workflow for MCNC benchmark circuits to the D-Wave Quantum Annealing architecture.

The hierarchy of this project is as follows:

Mapping

contains the flow for mapping the benchmark circuits to a simplified library and logic for generation of placement input files

Placing

contains a genetic algorithm for global cell placement and results of a simulated annealer placer https://github.com/kahlanlg/CADplacement, and the genetic algorithm with and without congestion awareness. Also includes logic for generation of detailed placement/routing input files.

Routing

contains a Lee-Moore routing algorithm based on a previous implementation https://github.com/kahlanlg/CADrouting. Also contains results from this algorithm and logic for generation of output files for placement image generation.

images

contains logic to generate an output image from the routed result, and images of results obtained for this project.

benchmarks

contains the unmapped MCNC benchmark circuits

demo

contains a simple circuit example

Execution Instructions

1. clone the repo into your local ABC installation
2. Perform automatic mapping, placing, routing, and image generation by executing bash run.sh

Dependencies/Credit

1. ABC http://people.eecs.berkeley.edu/~alanmi/abc/
2. MCNC Benchmark Circuits https://people.engr.ncsu.edu/brglez/
3. Python 2.7