Large-scale benchmarks of the Time-Warp/Graph-Theoretical Kinetic Monte Carlo approach for distributed on-lattice simulations of catalytic kinetics
We extend the work of Ravipati et al.[Comput. Phys. Commun., 2022, 270, 108148] in benchmarking the performance of large-scale, distributed, on-lattice kinetic Monte Carlo (KMC) simulations. Our software package, Zacros, employs a graph-theoretical approach to KMC, coupled with the Time-Warp algorithm for parallel discrete event simulations. The lattice is divided into equal subdomains, each assigned to a single processor; the cornerstone of the Time-Warp algorithm is the state queue, to which snapshots of the KMC (lattice) state are saved regularly, enabling historical KMC information to be corrected when conflicts occur at the subdomain boundaries. Focusing on three model systems, we highlight the key Time-Warp parameters that can be tuned to optimise KMC performance. The frequency of state saving, controlled by the state saving interval, δsnap, is shown to have the largest effect on performance, which favours balancing the overhead of re-simulating KMC history with that of writing state snapshots to memory. Also important is the global virtual time (GVT) computation interval, ΔτGVT, which has little direct effect on the progress of the simulation but controls how often the state queue memory can be freed up. We find that a vector data structure is, in general, more favourable than a linked list for storing the state queue, due to the reduced time required for allocating and de-allocating memory. These findings will guide users in maximising the efficiency of Zacros or other distributed KMC software, which is a vital step towards realising accurate, meso-scale simulations of heterogeneous catalysis.
- This article is part of the themed collection: Computational Modelling as a Tool in Catalytic Science