GooFit: A library for massively parallelising maximum-likelihood fits
University of Cincinnati, Physics Department, ML0011, Cincinnati OH 45221-0011, USA
arXiv:1311.1753 [cs.DC], (7 Nov 2013)
@article{2013arXiv1311.1753A,
author={Andreassen}, R. and {Meadows}, B.~T. and {de Silva}, M. and {Sokoloff}, M.~D. and {Tomko}, K.},
title={"{GooFit: A library for massively parallelising maximum-likelihood fits}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1311.1753},
primaryClass={"cs.DC"},
keywords={Computer Science – Distributed, Parallel, and Cluster Computing, Computer Science – Mathematical Software},
year={2013},
month={nov},
adsurl={http://adsabs.harvard.edu/abs/2013arXiv1311.1753A},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
Fitting complicated models to large datasets is a bottleneck of many analyses. We present GooFit, a library and tool for constructing arbitrarily-complex probability density functions (PDFs) to be evaluated on nVidia GPUs or on multicore CPUs using OpenMP. The massive parallelisation of dividing up event calculations between hundreds of processors can achieve speedups of factors 200-300 in real-world problems.
November 8, 2013 by hgpu
