Accelerating Trigger Performance of the ALICE Detector Using FPGA-Based Neural Networks

Abstract

The newly proposed heavy-ion detector at CERN Large Hadron Collider, ALICE3, will face a hundredfold increase in data rate due to the increased multiplicity and luminosity for both the proton-proton(pp) and lead-lead (Pb- Pb) collisions. Current CPU and GPU hardware combinations account for only 3.5 Gb/s, more than an order of magnitude lower than future demands. This research explores the use of machine learning algorithms on custom Field-Programmable Gate Arrays (FPGAs), a combination not yet