Speeding up complex multivariate data analysis in Borexino with parallel computing based on Graphics Processing Unit

TL;DR

This paper presents a GPU-based spectral fitting tool for Borexino that drastically reduces analysis time from hours or days to seconds or minutes, enhancing efficiency in solar neutrino data analysis.

Contribution

The development of a GPU-accelerated spectral fitter for Borexino that significantly speeds up complex data analysis compared to traditional CPU methods.

Findings

Fitting time reduced from hours/days to seconds/minutes.

GPU fitter outperforms CPU fitter in speed and efficiency.

Successfully integrated with Borexino analysis workflows.

Abstract

A spectral fitter based on the graphics processor unit (GPU) has been developed for Borexino solar neutrino analysis. It is able to shorten the fitting time to a superior level compared to the CPU fitting procedure. In Borexino solar neutrino spectral analysis, fitting usually requires around one hour to converge since it includes time-consuming convolutions in order to account for the detector response and pile-up effects. Moreover, the convergence time increases to more than two days when including extra computations for the discrimination of $^{11}$C and external $\gamma$s. In sharp contrast, with the GPU-based fitter it takes less than 10 seconds and less than four minutes, respectively. This fitter is developed utilizing the GooFit project with customized likelihoods, pdfs and infrastructures supporting certain analysis methods. In this proceeding the design of the package, developed features and the comparison with the original CPU fitter are presented.