Year: 2026
Type: article
Abstract: Simulation of detector response takes the largest amount of the computational resources of the ATLAS experiment at the LHC, with a large part used for simulation of calorimeters. As demand of CPU resources is expected to increase during Run 3 and in view of HL-LHC Run 4, action to ease this burden is required. One of the solutions developed by the Collaboration is the fast simulation system AtlFast3, which uses classical parametric and machine learning-based approaches to run simulation faster and with a smaller resource footprint with respect to Geant4. This work presents the version of AtlFast3 currently in production for simulation of Run 3 samples, discussing technical details, computing and physics performance, and the ongoing work in view of Run 4.
Author: Corchia, F. A. G.
Language: English
Subfield: Nuclear and High Energy Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7224972263