Lethe-DEM: an open-source parallel discrete element solver with load balancing

Approximately $${75}\%$$ of the raw material and $${50}\%$$ products in chemical industry are granular materials. The discrete element method (DEM) provides detailed insights phenomena at particle scale, it is therefore often used for modeling However, because DEM tracks motion contact individual particles separately, its computational cost increases nonlinearly $$O(n_\mathrm{p}\log (n_\mathrm{p}))$$ – $$O(n_\mathrm{p}^2)$$ (depending on algorithm) with number ( $$n_\mathrm{p}$$ ). In this article, we introduce a new open-source parallel software load balancing: Lethe-DEM. Lethe-DEM, module Lethe, consists solvers two-dimensional three-dimensional simulations. Load balancing allows Lethe-DEM to significantly increase efficiency by $$\approx {25}$$ $${70}\%$$ depending simulation. We explain fundamental modules architecture, governing equations. Furthermore, verify several tests including analytical solutions comparison other software. Comparisons experiments flat-bottomed silo, wedge-shaped rotating drum validate investigate strong weak scaling $${1}\le n_\mathrm{c} \le {192}$$ $${32}\le {320}$$ processes, respectively, without balancing. strong-scaling analysis performed silo simulations, while weak-scaling analysis, use dam-break best scalability obtained range $${5000}\le n_\mathrm{p}/n_\mathrm{c} {15{,}000}$$ . Finally, demonstrate that large-scale simulations can be carried out using simulation cylindrical $$n_\mathrm{p}={4.3}\times 10^6$$ 320 cores.