Chapter 20

Optimizing classical molecular dynamics in LAMMPS

Abstract

LAMMPS results on Knights Landing are discussed. Significant performance improvements on Intel Xeon processors and Knights Landing many-core processors are discussed. The simulations on single-socket Knights Landing processors were up to 1.96 × faster than dual socket Intel Xeon processors in configurations with a significantly lower processor TDP.

Keywords

Molecular dynamics; LAMMPS; SoA; AoS; MCDRAM; Numactl; AVX-512; Xeon Phi; Many-core; Knights Landing; AVX-512CD; Conflict detection instructions; Intrinsics; AVX-512 intrinsics; AVX-512CD intrinsics

Acknowledgment

Sandia is a multipurpose laboratory operated by Sandia Corporation, a Lockheed-Martin Co., for the US Department ...

Get Intel Xeon Phi Processor High Performance Programming, 2nd Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.