Parallelization of a Dynamic Monte Carlo Algorithm: A Partially Rejection-Free Conservative Approach

✍ Scribed by G Korniss; M.A Novotny; P.A Rikvold

• Publisher: Elsevier Science
• Year: 1999
• Tongue: English
• Weight: 415 KB
• Volume: 153
• Category: Article
• ISSN: 0021-9991
• DOI: 10.1006/jcph.1999.6291

✦ Synopsis

We experiment with a massively parallel implementation of an algorithm for simulating the dynamics of metastable decay in kinetic Ising models. The parallel scheme is directly applicable to a wide range of stochastic cellular automata where the discrete events (updates) are Poisson arrivals. For high performance, we utilize a continuoustime, asynchronous parallel version of the n-fold way rejection-free algorithm. Each processing element carries an l × l block of spins, and we employ fast one-sided communication routines on a distributed-memory parallel architecture. Different processing elements have different local simulated times. To ensure causality, the algorithm handles the asynchrony in a conservative fashion. Despite relatively low utilization and an intricate relationship between the average time increment and the size of the spin blocks, we find that the algorithm is scalable and for sufficiently large l it outperforms its corresponding parallel Metropolis (non-rejection-free) counterpart. As a sample application, we present results for metastable decay in a model ferromagnetic or ferroelectric film, observed with a probe of area smaller than the total system.