학술논문
Speeding up complex multivariate data analysis in Borexino with parallel computing based on Graphics Processing Unit
Document Type
Working Paper
Author
Ding, X. F.; Agostini, M.; Altenmuller, K.; Appel, S.; Atroshchenko, V.; Bagdasarian, Z.; Basilico, D.; Bellini, G.; Benziger, J.; Bick, D.; Bonfini, G.; Bravo, D.; Caccianiga, B.; Calaprice, F.; Caminata, A.; Caprioli, S.; Carlini, M.; Cavalcante, P.; Chepurnov, A.; Choi, K.; Collica, L.; D'Angelo, D.; Davini, S.; Derbin, A.; Di Ludovico, A.; Di Noto, L.; Drachnev, I.; Fomenko, K.; Formozov, A.; Franco, D.; Froborg, F.; Gabriele, F.; Galbiati, C.; Ghiano, C.; Giammarchi, M.; Goretti, A.; Gromov, M.; Guffanti, D.; Hagner, C.; Houdy, T.; Hungerford, E.; Ianni, Aldo; Ianni, Andrea; Jany, A.; Jeschke, D.; Kobychev, V.; Korablev, D.; Korga, G.; Kryn, D.; Laubenstein, M.; Litvinovich, E.; Lombardi, F.; Lombardi, P.; Ludhova, L.; Lukyanchenko, G.; Lukyanchenko, L.; Machulin, I.; Manuzio, G.; Marcocci, S.; Martyn, J.; Meroni, E.; Meyer, M.; Miramonti, L.; Misiaszek, M.; Muratova, V.; Neumair, B.; Oberauer, L.; Opitz, B.; Orekhov, V.; Ortica, F.; Pallavicini, M.; Papp, L.; Penek, O.; Pilipenko, N.; Pocar, A.; Porcelli, A.; Ranucci, G.; Razeto, A.; Re, A.; Redchuk, M.; Romani, A.; Roncin, R.; Rossi, N.; Schonert, S.; Semenov, D.; Skorokhvatov, M.; Smirnov, O.; Sotnikov, A.; Stokes, L. F. F.; Suvorov, Y.; Tartaglia, R.; Testera, G.; Thurn, J.; Toropova, M.; Unzhakov, E.; Vishneva, A.; Vogelaar, R. B.; von Feilitzsch, F.; Wang, H.; Weinz, S.; Wojcik, M.; Wurm, M.; Yokley, Z.; Zaimidoroga, O.; Zavatarelli, S.; Zuber, K.; Zuzel, G.
Source
Journal of Physics: Conference Series (2020)
Subject
Language
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.
Comment: 5 pages, 2 figures, proceeding for TAUP 2017 XV International Conference on Topics in Astroparticle and Underground Physics
Comment: 5 pages, 2 figures, proceeding for TAUP 2017 XV International Conference on Topics in Astroparticle and Underground Physics