Real-time Graph Building on FPGAs for Machine Learning Trigger Applications in Particle Physics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Neu, Marc, Becker, Juergen, Dorwarth, Philipp, Ferber, Torben, Reuter, Lea, Stefkova, Slavomira, Unger, Kai
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910376666333184
author Neu, Marc
Becker, Juergen
Dorwarth, Philipp
Ferber, Torben
Reuter, Lea
Stefkova, Slavomira
Unger, Kai
author_facet Neu, Marc
Becker, Juergen
Dorwarth, Philipp
Ferber, Torben
Reuter, Lea
Stefkova, Slavomira
Unger, Kai
contents We present a design methodology that enables the semi-automatic generation of a hardware-accelerated graph building architectures for locally constrained graphs based on formally described detector definitions. In addition, we define a similarity measure in order to compare our locally constrained graph building approaches with commonly used k-nearest neighbour building approaches. To demonstrate the feasibility of our solution for particle physics applications, we implemented a real-time graph building approach in a case study for the Belle~II central drift chamber using Field-Programmable Gate Arrays~(FPGAs). Our presented solution adheres to all throughput and latency constraints currently present in the hardware-based trigger of the Belle~II experiment. We achieve constant time complexity at the expense of linear space complexity and thus prove that our automated methodology generates online graph building designs suitable for a wide range of particle physics applications. By enabling an hardware-accelerated pre-processing of graphs, we enable the deployment of novel Graph Neural Networks~(GNNs) in first level triggers of particle physics experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2307_07289
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Real-time Graph Building on FPGAs for Machine Learning Trigger Applications in Particle Physics
Neu, Marc
Becker, Juergen
Dorwarth, Philipp
Ferber, Torben
Reuter, Lea
Stefkova, Slavomira
Unger, Kai
High Energy Physics - Experiment
Signal Processing
We present a design methodology that enables the semi-automatic generation of a hardware-accelerated graph building architectures for locally constrained graphs based on formally described detector definitions. In addition, we define a similarity measure in order to compare our locally constrained graph building approaches with commonly used k-nearest neighbour building approaches. To demonstrate the feasibility of our solution for particle physics applications, we implemented a real-time graph building approach in a case study for the Belle~II central drift chamber using Field-Programmable Gate Arrays~(FPGAs). Our presented solution adheres to all throughput and latency constraints currently present in the hardware-based trigger of the Belle~II experiment. We achieve constant time complexity at the expense of linear space complexity and thus prove that our automated methodology generates online graph building designs suitable for a wide range of particle physics applications. By enabling an hardware-accelerated pre-processing of graphs, we enable the deployment of novel Graph Neural Networks~(GNNs) in first level triggers of particle physics experiments.
title Real-time Graph Building on FPGAs for Machine Learning Trigger Applications in Particle Physics
topic High Energy Physics - Experiment
Signal Processing
url https://arxiv.org/abs/2307.07289