A GPU-Based Parallel Procedure for Nonlinear Analysis of Complex Structures Using a Coupled FEM/DEM Approach
Figure 11
Runtime comparison among different GPU platforms on different computation precisions. “Stiffness” means generating stiffness matrices; “Iteration” is short for dynamic relaxation iteration; “Elements” represents elements and nodal calculation; “Cloning” denotes nodal force redistribution; “Energy” stands for energy calculation; “Output” refers to data output.
