Implementing noise reduction techniques into theOpenQ*D package

TL;DR

This paper tests a new stochastic noise reduction technique in the OpenQ*D package for lattice QCD calculations, finding it does not improve speed due to spectral differences in operators.

Contribution

The paper implements and evaluates a novel noise reduction method within OpenQ*D for Wilson fermion ensembles, highlighting its limitations.

Findings

No speedup achieved with the new technique

Spectral differences hinder the technique's effectiveness

Method's assumption about spectra similarity is invalid in tested cases

Abstract

We present the results of testing a new technique for stochastic noise reduction in the calculation of propagators by implementing it in OpenQ*D for two ensembles with O(a) improved Wilson fermion action, with periodic boundary conditions and pion masses of 437 MeV and 331 MeV, for the connected vector and pseudoscalar correlators. We find that the technique yields no speedup compared to traditional methods, owning to the failure of its underlying assumption that the spectra of the spatial Laplacian and Dirac operators are sufficiently similar for the technique's purposes.