TweTriS: Twenty trillion-atom simulation. (September 2019)

• Record Type: Journal Article
• Authors: Tchipev, Nikola; Seckler, Steffen; Heinen, Matthias; Vrabec, Jadran; Gratl, Fabio; Horsch, Martin; Bernreuther, Martin; Glass, Colin W; Niethammer, Christoph; Hammer, Nicolay; Krischok, Bernd; Resch, Michael; Kranzlmüller, Dieter; Hasse, Hans; Bungartz, Hans-Joachim; Neumann, Philipp
• Other Names: Mascagni Michael guest-editor.
• Is Part Of: International journal of high performance computing applications. Volume 33:Number 5(2019)
• Page Start: 838
• Page End: 854
• Publication Date: 2019-09
• Subjects: Molecular dynamics -- SIMD vectorization -- OpenMP -- MPI -- coloring -- memory-efficient -- SuperMUC -- Hazel Hen; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342018819741
• Languages: English
• ISSNs: 1094-3420

Abstract

Significant improvements are presented for the molecular dynamics codels1 mardyn — a linked cell-based code for simulating a large number of small, rigid molecules with application areas in chemical engineering. The changes consist of a redesign of the SIMD vectorization via wrappers, MPI improvements and a software redesign to allow memory-efficient execution with the production trunk to increase portability and extensibility. Two novel, memory-efficient OpenMP schemes for the linked cell-based force calculation are presented, which are able to retain Newton's third law optimization. Comparisons to well-optimized Verlet list-based codes, such as LAMMPS and GROMACS, demonstrate the viability of the linked cell-based approach. The present version ofls1 mardyn is used to run simulations on entire supercomputers, maximizing the number of sampled atoms. Compared to the preceding version ofls1 mardyn on the entire set of 9216 nodes of SuperMUC, Phase 1, 27% more atoms are simulated. Weak scaling performance is increased by up to 40% and strong scaling performance by up to more than 220%. On Hazel Hen, strong scaling efficiency of up to 81% and 189 billion molecule updates per second is attained, when scaling from 8 to 7168 nodes. Moreover, a total of 20 trillion atoms is simulated at up to 88% weak scaling efficiency running at up to 1.33 PFLOPS. This represents a fivefold increase in terms of the number of atoms simulated to date.