Multipath Metropolis simulation: An application to the classical Heisenberg model

• Multipath Metropolis simulation of classical Heisenberg model is explored.
• Multipath simulation approach is inherently parallel because different random-walk paths are mutually independent.
• This independence enables easy and efficient harnessing of numerous cores’ computing power in embarrassingly parallel algorithms.

This study explores the Multipath Metropolis simulation of the classical Heisenberg model. Unlike the standard single-path algorithm, the Metropolis algorithm applied to multiple random-walk paths becomes an embarrassingly parallel algorithm in which many processor cores can be easily utilized. This is important since processor cores are progressively becoming less expensive and thus more accessible. The most obvious advantage of the multipath approach is in employing independent random-walk paths to produce an uncorrelated simulation output with a normal distribution allowing for straightforward and rigorous statistical analysis.