Open AccessANALYSIS OF THE EFFECTIVENESS OF VARIOUS METHODS FOR PARALLELIZING DATA PROCESSING IMPLEMENTED IN THE ROOT PACKAGE
Author(s) -
T.M. Solovjeva
Publication year - 2021
Publication title -
9th international conference "distributed computing and grid technologies in science and education"
Language(s) - English
Resource type - Conference proceedings
DOI - 10.54546/mlit.2021.78.10.002
Subject(s) - computer science , merge (version control) , scalability , parallel computing , network packet , structuring , speedup , operating system , computer network , finance , economics
The ROOT software package is currently being upgraded in several ways to improve data processingperformance. This paper will consider several tools implemented in this framework for calculations onmodern heterogeneous computing architectures. PROOF (Parallel ROOT Package Extension) dividescommon work into small chunks, i.e. packets. The size of the first packet used for calibration, theminimum and maximum set size of the packet, the degree of data structuring affect the speed of theirprocessing. When processing large amounts of data, the read and write speed can be crucial. The newasynchronous file merge feature in the TBufferMerger class allows writing data in parallel frommultiple streams to a single output file. Our calculations show a good scalability of the macroexecution time depending on the number of processor cores used.