Graph Clustering: a graph-based clustering algorithm for the electromagnetic calorimeter in LHCb

Abstract The recent upgrade of the LHCb experiment pushes data processing rates up to 40 Tbit/s. Out whole reconstruction sequence, one most time consuming algorithms is calorimeter reconstruction. It aims at performing a clustering readout cells from detector that belong same particle in order measure its energy and position. This article presents new algorithm for makes use graph structures optimise process, will be denoted Graph Clustering. outperforms previously used method by $$65.4\%$$ 65.4 % terms computational on average, with an equivalent efficiency resolution. implementation Clustering detailed this article, together performance results inside framework using simulation data.