Data di Pubblicazione:
2021
Citazione:
PDFFlow: Parton distribution functions on GPU / S. Carrazza, C. Juan, M. Rossi. - In: COMPUTER PHYSICS COMMUNICATIONS. - ISSN 0010-4655. - 264(2021 Jul). [10.1016/j.cpc.2021.107995]
Abstract:
We present PDFFlow, a new software for fast evaluation of parton distribution functions (PDFs) designed for platforms with hardware accelerators. PDFs are essential for the calculation of particle physics observables through Monte Carlo simulation techniques. The evaluation of a generic set of PDFs for quarks and gluon at a given momentum fraction and energy scale requires the implementation of interpolation algorithms as introduced for the first time by the LHAPDF project. PDFFlow extends and implements these interpolation algorithms using Google's TensorFlow library providing the capabilities to perform PDF evaluations taking fully advantage of multi-threading CPU and GPU setups. We benchmark the performance of this library on multiple scenarios relevant for the particle physics community. Program summary: Program Title: PDFFlow CPC Library link to program files: https://doi.org/10.17632/rtp8xr3hn9.1 Developer's repository link: https://github.com/N3PDF/pdfflow Licensing provisions: GPLv3 Programming language: Python, C Nature of problem: The evaluation of a generic set of parton distribution functions requires the implementation of interpolation algorithms. Currently, there are no public available implementations with hardware acceleration support. Solution method: Implementation of interpolation algorithms for the evaluation of parton distribution functions and the strong coupling αs using the dataflow graph infrastructure provided by the TensorFlow framework, taking advantage of multi-threading CPU and GPU setups.
Tipologia IRIS:
01 - Articolo su periodico
Keywords:
Graphs; Hardware acceleration; Machine learning; Parton distributions;
Elenco autori:
S. Carrazza, C. Juan, M. Rossi
Link alla scheda completa:
Link al Full Text: