GPU coprocessors as a service for deep learning inference in high energy physics
| Autor: | Jeffrey Krupa, Maria Acosta Flechas, Kelvin Lin, Jack Dinsmore, Javier Duarte, Scott Hauck, Nhan Tran, Burt Holzman, Natchanon Suaysom, Mia Liu, Thomas Klijnsma, Kevin Pedro, Philip Harris, Dylan Rankin, Matthew Trahms, Shih-Chieh Hsu |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Particle physics Physics - Instrumentation and Detectors Coprocessor FOS: Physical sciences 01 natural sciences 7. Clean energy High Energy Physics - Experiment law.invention High Energy Physics - Experiment (hep-ex) Artificial Intelligence law 0103 physical sciences Graphics 010306 general physics Collider Large Hadron Collider Computer performance 010308 nuclear & particles physics business.industry Deep learning Instrumentation and Detectors (physics.ins-det) Computational Physics (physics.comp-ph) 3. Good health Human-Computer Interaction Workflow Computer Science - Distributed Parallel and Cluster Computing Physics - Data Analysis Statistics and Probability Hardware acceleration Distributed Parallel and Cluster Computing (cs.DC) Artificial intelligence business Physics - Computational Physics Data Analysis Statistics and Probability (physics.data-an) Software |
| Zdroj: | Machine Learning: Science and Technology. 2:035005 |
| ISSN: | 2632-2153 |
| DOI: | 10.1088/2632-2153/abec21 |
| Popis: | In the next decade, the demands for computing in large scientific experiments are expected to grow tremendously. During the same time period, CPU performance increases will be limited. At the CERN Large Hadron Collider (LHC), these two issues will confront one another as the collider is upgraded for high luminosity running. Alternative processors such as graphics processing units (GPUs) can resolve this confrontation provided that algorithms can be sufficiently accelerated. In many cases, algorithmic speedups are found to be largest through the adoption of deep learning algorithms. We present a comprehensive exploration of the use of GPU-based hardware acceleration for deep learning inference within the data reconstruction workflow of high energy physics. We present several realistic examples and discuss a strategy for the seamless integration of coprocessors so that the LHC can maintain, if not exceed, its current performance throughout its running. 26 pages, 7 figures, 2 tables |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|