Investigation of heterogeneous computing platforms for real-time data analysis in the CBM experiment

TL;DR

This paper explores how heterogeneous computing platforms like GPUs and CPUs can be effectively used for real-time data analysis in high-energy physics experiments, specifically for the CBM experiment's online event selection.

Contribution

It compares various parallel computing paradigms on CPUs and GPUs, demonstrating the feasibility of real-time data processing with optimized hardware deployment.

Findings

OpenCL can serve as a unified parallel paradigm across hardware.

Heterogeneous platforms can handle high data rates with moderate resources.

Optimized use of compute power is crucial for real-time analysis.

Abstract

Future experiments in high-energy physics will pose stringent requirements to computing, in particular to real-time data processing. As an example, the CBM experiment at FAIR Germany intends to perform online data selection exclusively in software, without using any hardware trigger, at extreme interaction rates of up to 10 MHz. In this article, we describe how heterogeneous computing platforms, Graphical Processing Units (GPUs) and CPUs, can be used to solve the associated computing problems on the example of the first-level event selection process sensitive to J/{\psi} decays using muon detectors. We investigate and compare pure parallel computing paradigms (Posix Thread, OpenMP, MPI) and heterogeneous parallel computing paradigms (CUDA, OpenCL) on both CPU and GPU architectures and demonstrate that the problem under consideration can be accommodated with a moderate deployment of hardware resources, provided their compute power is made optimal use of. In addition, we compare OpenCL and pure parallel computing paradigms on CPUs and show that OpenCL can be considered as a single parallel paradigm for all hardware resources.