Fast DEM collision checks on multicore nodes

Krestenitis, Konstantinos; Weinzierl, Tobias; Koziara, Tomasz

Abstract

Many particle simulations today rely on spherical or analytical particle shape descriptions. They find non-spherical, triangulated particle models computationally infeasible due to expensive collision detections. We propose a hybrid collision detection algorithm based upon an iterative solve of a minimisation problem that automatically falls back to a brute-force comparison-based algorithm variant if the problem is ill-posed. Such a hybrid can exploit the vector facilities of modern chips and it is well-prepared for the arising manycore era. Our approach pushes the boundary where non-analytical particle shapes and the aligning of more accurate first principle physics become manageable.

Citation

Krestenitis, K., Weinzierl, T., & Koziara, T. (2017, September). Fast DEM collision checks on multicore nodes. Presented at 12th International Conference on Parallel Processing and Applied Mathematics (PPAM) 2017, Lublin, Poland

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 12th International Conference on Parallel Processing and Applied Mathematics (PPAM) 2017
• Start Date: Sep 10, 2017
• End Date: Sep 13, 2017
• Acceptance Date: Sep 4, 2017
• Online Publication Date: Mar 23, 2018
• Publication Date: Mar 23, 2018
• Deposit Date: Sep 4, 2017
• Publicly Available Date: Mar 23, 2019
• Print ISSN: 0302-9743
• Pages: 123-132
• Series Title: Lecture notes in computer science
• Series Number: 10777
• Series ISSN: 0302-9743,1611-3349
• Book Title: Parallel processing and applied mathematics : 12th International conference, PPAM 2017, Lublin, Poland, September 10-13 ; revised selected papers. Part 1.
• ISBN: 9783319780238
• DOI: https://doi.org/10.1007/978-3-319-78024-5_12
• Public URL: https://durham-repository.worktribe.com/output/1146453