Graph Neural Network-based Tracking as a Service

Year: 2024

Type: preprint

Abstract: Recent studies have shown promising results for track finding in dense environments using Graph Neural Network (GNN)-based algorithms. However, GNN-based track finding is computationally slow on CPUs, necessitating the use of coprocessors to accelerate the inference time. Additionally, the large input graph size demands a large device memory for efficient computation, a requirement not met by all computing facilities used for particle physics experiments, particularly those lacking advanced GPUs. Furthermore, deploying the GNN-based track-finding algorithm in a production environment requires the installation of all dependent software packages, exclusively utilized by this algorithm. These computing challenges must be addressed for the successful implementation of GNN-based track-finding algorithm into production settings. In response, we introduce a ``GNN-based tracking as a service'' approach, incorporating a custom backend within the NVIDIA Triton inference server to facilitate GNN-based tracking. This paper presents the performance of this approach using the Perlmutter supercomputer at NERSC.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2402.09633

Authors: Zhao, Haoran, Naylor, Andrew, Hsu, Shih-Chieh, Calafiura, Paolo, Farrell, Steven, Feng, Yongbing, Harris, Philip Coleman, Khoda, Elham E, Mccormack, William Patrick, Rankin, Dylan Sheldon, Ju, Xiangyang

Institutions: University of Washington, National Energy Research Scientific Computing Center, Lawrence Berkeley National Laboratory, Fermi National Accelerator Laboratory, Massachusetts Institute of Technology, University of Pennsylvania

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, U.S. Department of Energy, National Energy Research Scientific Computing Center, Office of Science, National Nuclear Security Administration, High Energy Physics, Lawrence Berkeley National Laboratory

OpenAlex ID: W4391912562