12–13 Dec 2023
DESY Hamburg and Zoom
Europe/Berlin timezone

columnflow: Fully automated, distributed framework for vectorized Run 3 analyses

12 Dec 2023, 14:00
20m
Main Auditorium (DESY Hamburg)

Main Auditorium

DESY Hamburg

Computing and Machine Learning ML+computing Parallel

Speaker

Nathan Prouvost (UNI/EXP (Uni Hamburg, Institut fur Experimentalphysik))

Description

In order to observe and measure rare processes in nature, a staggering amount of data needs to be produced and processed at particle colliders. With the advancement of the LHC towards Run 3 and HL-LHC, the flow of data as well as the complexity of the analyses will increase even more. In light of these challenges and the limited resources available, an efficient usage of computing power is critical for future analyses.
In order to analyze data in an efficient way, a new columnar analysis tool, columnflow, has been developed. In this presentation, an introduction to columnflow is given, including an overview of the workflow and some example use cases.

Primary authors

Marcel Rieger (UNI/EXP (Uni Hamburg, Institut fur Experimentalphysik)) Ana Rita Alves Andrade (None) Bogdan Wiederspan (UNI/EXP (Uni Hamburg, Institut fur Experimentalphysik)) Nathan Prouvost (UNI/EXP (Uni Hamburg, Institut fur Experimentalphysik)) Peter Schleper (UNI/EXP (Uni Hamburg, Institut fur Experimentalphysik)) Philip Daniel Keicher (UNI/EXP (Uni Hamburg, Institut fur Experimentalphysik))

Presentation materials