A Note on Time Measurements in LAMMPS

TL;DR

This paper critically examines how LAMMPS measures component efficiency, identifies inaccuracies in its current approach, and proposes correct methods and code modifications for accurate performance assessment.

Contribution

It reveals flaws in LAMMPS's current efficiency measurement approach and provides a corrected methodology along with necessary code changes.

Findings

Current LAMMPS approach can lead to misleading efficiency conclusions

Experimental verification confirms inaccuracies in existing measurements

Proposed code modifications enable correct efficiency assessment

Abstract

We examine the issue of assessing the efficiency of components of a parallel program at the example of the MD package LAMMPS. In particular, we look at how LAMMPS deals with the issue and explain why the approach adopted might lead to inaccurate conclusions. The misleading nature of this approach is subsequently verified experimentally with a case study. Afterwards, we demonstrate how one should correctly determine the efficiency of the components and show what changes to the code base of LAMMPS are necessary in order to get the correct behavior.