Bulletin of the American Physical Society
APS April Meeting 2024
Wednesday–Saturday, April 3–6, 2024; Sacramento & Virtual
Session DD03: V: Data Science and AI/ML in Physics
5:30 AM–6:30 AM,
Thursday, April 4, 2024
Room: Virtual Room 03
Sponsoring
Unit:
GDS
Chair: Jonte Hance, Newcastle University
Abstract: DD03.00005 : Graph Neural Network-based Track finding as a Service with ACTS*
6:18 AM–6:30 AM
Presenter:
Haoran Zhao
(University of Washington)
Authors:
Yuan-Tang Chou
(University of Washington)
Haoran Zhao
(University of Washington)
Xiangyang Ju
(Lawrence Berkeley National Laboratory)
Shih-Chieh Hsu
(University of Washington)
Paolo Calafiura
(Lawrence Berkeley National Laboratory)
Philip C Harris
(Massachusetts Institute of Technology)
Patrick McCormack
(Massachusetts Institute of Technology)
Yao Yao
(Purdue University)
Yongbin Feng
(Fermi National Accelerator Laboratory)
Elham E Khoda
(University of Washington)
Kevin J Pedro
(Fermi National Accelerator Laboratory)
Dylan S Rankin
(University of Pennsylvania)
Andrew Naylor
(Lawrence Berkeley National Laboratory)
To address this, we propose deploying the GNN-based track-finding algorithm as a service in the cloud or high-performance computing centers such as the NERSC Perlmutter system with over 7000 A100 GPUs. We have implemented a tracking-as-a-service prototype within A Common Tracking Software (ACTS), a toolkit for charged particle track reconstruction.
This approach is algorithm-agnostic, allowing the incorporation of various algorithms as new backends through interactions with the client interface in ACTS. In this contribution, we showcase the versatility of the as-a-service approach by implementing the GNN-based track-finding workflow using the Nvidia Triton Inference Server within ACTS. We assess track-finding throughput and GPU utilization, exploring the scalability of the inference server across the NERSC Perlmutter supercomputer and cloud resources.
*This research used the National Energy Research Scientific Computing Center (NERSC) resources, a U.S. Department of Energy Office of Science User Facility located at Lawrence Berkeley National Laboratory, operated under Contract No. DE-AC02-05CH11231 using NERSC award ERCAP0021226 and is supported by NSF award No. 2117997
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700