Scaling the SIESTA Magnetohydrodynamics Equilibrium Code

Abstract

We report the results of a scaling effort that increases both the speed and resolution of the SIESTA magnetohydrodynamics equilibrium code. SIESTA is capable of computing three-dimensional plasma equilibria with magnetic islands at high spatial resolutions for toroidally confined plasmas. Starting with a small-scale parallel implementation, we identified scale-dependent bottlenecks of the code and developed scalable alternatives for each performance-significant functionality, cumulatively improving both its runtime speed (on the same number of processors) and its scalability (across larger number of processors) by an order of magnitude. The net outcome is an improvement in speed by over 10-fold, utilizing a few thousand processors, enabling SIESTA to simulate high spatial-resolution scenarios in under an hour for the first time.