Monte-Carlo Simulator for Polymerisation Kinetics

We developed a new method to improve the performance of simulators based on Monte-Carlo methods and used it to implement a simulator for polymerisation kinetics that includes support for star polymers. Our performance advantage is due to combining code specialisation with parallelisation (suitable for PC clusters, SMP, and multicore hardware).

Papers and source code

Two paper describing our approach to high-performance Monte-Carlo simulations are available (one aimed at a computing audience and one aimed at chemists) as well as Moreover, the source code of the simulator generator is available.

• Specialising Simulator Generators for High-Performance Monte-Carlo Methods, submitted for publication.
• A Parallelised High Performance Monte Carlo Simulation Approach for Complex Polymerisation Kinetics, Macromolecular Theory & Simulation 16(6).
• parapolysim-0.1.tar.gz (source code)

NB: The above version of the simulator generator does not read the reaction system from an .rsy file, but includes it as a Haskell data structure. This makes it easier to use the generator without access to PREDICI.

Support

This work was funded by the UNSW FRGP High Performance Parallel Computing for Complex Polymer Architecture Design.

People (alphabetically)

• Christopher Barner-Kowollik, University of New South Wales, Centre for Advanced Macromolecular Design (CAMD)
• Hugh Chaffey-Millar, University of New South Wales, Centre for Advanced Macromolecular Design (CAMD)
• Manuel Chakravarty, University of New South Wales, Programming Languages and Systems (PLS)
• Gabriele Keller, University of New South Wales, Programming Languages and Systems (PLS)
• Don Stewart, University of New South Wales, Programming Languages and Systems (PLS)