# GPU-based Online Track Reconstruction for the ALICE TPC in Run 3 with Continuous Read-Out

## TL;DR

This paper presents a GPU-accelerated online track reconstruction algorithm for the ALICE TPC during LHC Run 3, capable of processing 50 times more collisions per second with performance comparable to offline methods.

## Contribution

It introduces a novel GPU-based tracking algorithm using Cellular Automaton and Kalman filter for real-time data processing in high-rate collision environments.

## Key findings

- Achieved high compression ratios for raw data
- Demonstrated real-time processing on GPUs with performance comparable to offline reconstruction
- Successfully handled continuous read-out and complex collision scenarios

## Abstract

In LHC Run 3, ALICE will increase the data taking rate significantly to 50 kHz continuous read-out of minimum bias Pb-Pb collisions. The reconstruction strategy of the online-offline computing upgrade foresees a first synchronous online reconstruction stage during data taking enabling detector calibration and data compression, and a posterior calibrated asynchronous reconstruction stage. Many new challenges arise, among them continuous TPC read-out, more overlapping collisions, no a priori knowledge of the primary vertex and of location-dependent calibration in the synchronous phase, identification of low-momentum looping tracks, and sophisticated raw data compression. The tracking algorithm for the Time Projection Chamber (TPC) will be based on a Cellular Automaton and the Kalman filter. The reconstruction shall run online, processing 50 times more collisions per second than today, while yielding results comparable to current offline reconstruction. Our TPC track finding leverages the potential of hardware accelerators via the OpenCL and CUDA APIs in a shared source code for CPUs and GPUs for both reconstruction stages. We give an overview of the status of Run 3 tracking including performance on processors and GPUs and achieved compression ratios.