Accelerating scientific codes by performance and accuracy modeling

TL;DR

This paper introduces an automated parameter selection methodology for scientific simulation codes, optimizing accuracy and performance, demonstrated on molecular dynamics software, leading to significant reductions in execution time.

Contribution

The paper presents a general methodology and prototype tool for automatic parameter tuning in scientific codes, improving efficiency while maintaining accuracy.

Findings

Achieved 10-60% reduction in simulation time.

Enabled up to twice as many simulations within fixed core-hours.

Validated methodology on LAMMPS molecular dynamics software.

Abstract

Scientific software is often driven by multiple parameters that affect both accuracy and performance. Since finding the optimal configuration of these parameters is a highly complex task, it extremely common that the software is used suboptimally. In a typical scenario, accuracy requirements are imposed, and attained through suboptimal performance. In this paper, we present a methodology for the automatic selection of parameters for simulation codes, and a corresponding prototype tool. To be amenable to our methodology, the target code must expose the parameters affecting accuracy and performance, and there must be formulas available for error bounds and computational complexity of the underlying methods. As a case study, we consider the particle-particle particle-mesh method (PPPM) from the LAMMPS suite for molecular dynamics, and use our tool to identify configurations of the input parameters that achieve a given accuracy in the shortest execution time. When compared with the configurations suggested by expert users, the parameters selected by our tool yield reductions in the time-to-solution ranging between 10% and 60%. In other words, for the typical scenario where a fixed number of core-hours are granted and simulations of a fixed number of timesteps are to be run, usage of our tool may allow up to twice as many simulations. While we develop our ideas using LAMMPS as computational framework and use the PPPM method for dispersion as case study, the methodology is general and valid for a range of software tools and methods.