Cargando…
Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics
Graph neural networks have been shown to achieve excellent performance for several crucial tasks in particle physics, such as charged particle tracking, jet tagging, and clustering. An important domain for the application of these networks is the FGPA-based first layer of real-time data filtering at...
Autores principales: | , , , , , , , , , , , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8006281/ https://www.ncbi.nlm.nih.gov/pubmed/33791596 http://dx.doi.org/10.3389/fdata.2020.598927 |
_version_ | 1783672281892913152 |
---|---|
author | Iiyama, Yutaro Cerminara, Gianluca Gupta, Abhijay Kieseler, Jan Loncar, Vladimir Pierini, Maurizio Qasim, Shah Rukh Rieger, Marcel Summers, Sioni Van Onsem, Gerrit Wozniak, Kinga Anna Ngadiuba, Jennifer Di Guglielmo, Giuseppe Duarte, Javier Harris, Philip Rankin, Dylan Jindariani, Sergo Liu, Mia Pedro, Kevin Tran, Nhan Kreinar, Edward Wu, Zhenbin |
author_facet | Iiyama, Yutaro Cerminara, Gianluca Gupta, Abhijay Kieseler, Jan Loncar, Vladimir Pierini, Maurizio Qasim, Shah Rukh Rieger, Marcel Summers, Sioni Van Onsem, Gerrit Wozniak, Kinga Anna Ngadiuba, Jennifer Di Guglielmo, Giuseppe Duarte, Javier Harris, Philip Rankin, Dylan Jindariani, Sergo Liu, Mia Pedro, Kevin Tran, Nhan Kreinar, Edward Wu, Zhenbin |
author_sort | Iiyama, Yutaro |
collection | PubMed |
description | Graph neural networks have been shown to achieve excellent performance for several crucial tasks in particle physics, such as charged particle tracking, jet tagging, and clustering. An important domain for the application of these networks is the FGPA-based first layer of real-time data filtering at the CERN Large Hadron Collider, which has strict latency and resource constraints. We discuss how to design distance-weighted graph networks that can be executed with a latency of less than one μs on an FPGA. To do so, we consider a representative task associated to particle reconstruction and identification in a next-generation calorimeter operating at a particle collider. We use a graph network architecture developed for such purposes, and apply additional simplifications to match the computing constraints of Level-1 trigger systems, including weight quantization. Using the hls4ml library, we convert the compressed models into firmware to be implemented on an FPGA. Performance of the synthesized models is presented both in terms of inference accuracy and resource usage. |
format | Online Article Text |
id | pubmed-8006281 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-80062812021-03-30 Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics Iiyama, Yutaro Cerminara, Gianluca Gupta, Abhijay Kieseler, Jan Loncar, Vladimir Pierini, Maurizio Qasim, Shah Rukh Rieger, Marcel Summers, Sioni Van Onsem, Gerrit Wozniak, Kinga Anna Ngadiuba, Jennifer Di Guglielmo, Giuseppe Duarte, Javier Harris, Philip Rankin, Dylan Jindariani, Sergo Liu, Mia Pedro, Kevin Tran, Nhan Kreinar, Edward Wu, Zhenbin Front Big Data Big Data Graph neural networks have been shown to achieve excellent performance for several crucial tasks in particle physics, such as charged particle tracking, jet tagging, and clustering. An important domain for the application of these networks is the FGPA-based first layer of real-time data filtering at the CERN Large Hadron Collider, which has strict latency and resource constraints. We discuss how to design distance-weighted graph networks that can be executed with a latency of less than one μs on an FPGA. To do so, we consider a representative task associated to particle reconstruction and identification in a next-generation calorimeter operating at a particle collider. We use a graph network architecture developed for such purposes, and apply additional simplifications to match the computing constraints of Level-1 trigger systems, including weight quantization. Using the hls4ml library, we convert the compressed models into firmware to be implemented on an FPGA. Performance of the synthesized models is presented both in terms of inference accuracy and resource usage. Frontiers Media S.A. 2021-01-12 /pmc/articles/PMC8006281/ /pubmed/33791596 http://dx.doi.org/10.3389/fdata.2020.598927 Text en Copyright © 2021 Iiyama, Cerminara, Gupta, Kieseler, Loncar, Pierini, Qasim, Rieger, Summers, Van Onsem, Wozniak, Ngadiuba, Di Guglielmo, Duarte, Harris, Rankin, Jindariani, Liu, Pedro, Tran, Kreinar and Wu. http://Creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Big Data Iiyama, Yutaro Cerminara, Gianluca Gupta, Abhijay Kieseler, Jan Loncar, Vladimir Pierini, Maurizio Qasim, Shah Rukh Rieger, Marcel Summers, Sioni Van Onsem, Gerrit Wozniak, Kinga Anna Ngadiuba, Jennifer Di Guglielmo, Giuseppe Duarte, Javier Harris, Philip Rankin, Dylan Jindariani, Sergo Liu, Mia Pedro, Kevin Tran, Nhan Kreinar, Edward Wu, Zhenbin Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics |
title | Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics |
title_full | Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics |
title_fullStr | Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics |
title_full_unstemmed | Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics |
title_short | Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics |
title_sort | distance-weighted graph neural networks on fpgas for real-time particle reconstruction in high energy physics |
topic | Big Data |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8006281/ https://www.ncbi.nlm.nih.gov/pubmed/33791596 http://dx.doi.org/10.3389/fdata.2020.598927 |
work_keys_str_mv | AT iiyamayutaro distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT cerminaragianluca distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT guptaabhijay distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT kieselerjan distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT loncarvladimir distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT pierinimaurizio distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT qasimshahrukh distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT riegermarcel distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT summerssioni distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT vanonsemgerrit distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT wozniakkingaanna distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT ngadiubajennifer distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT diguglielmogiuseppe distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT duartejavier distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT harrisphilip distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT rankindylan distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT jindarianisergo distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT liumia distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT pedrokevin distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT trannhan distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT kreinaredward distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics AT wuzhenbin distanceweightedgraphneuralnetworksonfpgasforrealtimeparticlereconstructioninhighenergyphysics |