Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics
Machine protection is a core task of real-time image diagnostics aiming for steady-state operation in nuclear fusion devices. The paper evaluates the applicability of the newest low-power NVIDIA Jetson Xavier NX platform for image plasma diagnostics. This embedded NVIDIA Tegra System-on-a-Chip (SoC)...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-03-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/15/6/2088 |
_version_ | 1797471788969492480 |
---|---|
author | Bartłomiej Jabłoński Dariusz Makowski Piotr Perek Patryk Nowak vel Nowakowski Aleix Puig Sitjes Marcin Jakubowski Yu Gao Axel Winter The W-X Team |
author_facet | Bartłomiej Jabłoński Dariusz Makowski Piotr Perek Patryk Nowak vel Nowakowski Aleix Puig Sitjes Marcin Jakubowski Yu Gao Axel Winter The W-X Team |
author_sort | Bartłomiej Jabłoński |
collection | DOAJ |
description | Machine protection is a core task of real-time image diagnostics aiming for steady-state operation in nuclear fusion devices. The paper evaluates the applicability of the newest low-power NVIDIA Jetson Xavier NX platform for image plasma diagnostics. This embedded NVIDIA Tegra System-on-a-Chip (SoC) integrates a Graphics Processing Unit (GPU) and Central Processing Unit (CPU) on a single chip. The hardware differences and features compared to the previous NVIDIA Jetson TX2 are signified. Implemented algorithms detect thermal events in real-time, utilising the high parallelism provided by the embedded General-Purpose computing on Graphics Processing Units (GPGPU). The performance and accuracy are evaluated on the experimental data from the Wendelstein 7-X (W7-X) stellarator. Strike-line and reflection events are primarily investigated, yet benchmarks for overload hotspots, surface layers and visualisation algorithms are also included. Their detection might allow for automating real-time risk evaluation incorporated in the divertor protection system in W7-X. For the first time, the paper demonstrates the feasibility of complex real-time image processing in nuclear fusion applications on low-power embedded devices. Moreover, GPU-accelerated reference processing pipelines yielding higher accuracy compared to the literature results are proposed, and remarkable performance improvement resulting from the upgrade to the Xavier NX platform is attained. |
first_indexed | 2024-03-09T19:53:02Z |
format | Article |
id | doaj.art-54630e3567c54d8b92108d53fd72c5cb |
institution | Directory Open Access Journal |
issn | 1996-1073 |
language | English |
last_indexed | 2024-03-09T19:53:02Z |
publishDate | 2022-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj.art-54630e3567c54d8b92108d53fd72c5cb2023-11-24T01:04:19ZengMDPI AGEnergies1996-10732022-03-01156208810.3390/en15062088Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma DiagnosticsBartłomiej Jabłoński0Dariusz Makowski1Piotr Perek2Patryk Nowak vel Nowakowski3Aleix Puig Sitjes4Marcin Jakubowski5Yu Gao6Axel Winter7The W-X TeamDepartment of Microelectronics and Computer Science, Lodz University of Technology, 90-924 Łódź, PolandDepartment of Microelectronics and Computer Science, Lodz University of Technology, 90-924 Łódź, PolandDepartment of Microelectronics and Computer Science, Lodz University of Technology, 90-924 Łódź, PolandDepartment of Microelectronics and Computer Science, Lodz University of Technology, 90-924 Łódź, PolandStellarator Edge and Divertor Physics Division, Max Planck Institute for Plasma Physics, 17491 Greifswald, GermanyStellarator Edge and Divertor Physics Division, Max Planck Institute for Plasma Physics, 17491 Greifswald, GermanyStellarator Edge and Divertor Physics Division, Max Planck Institute for Plasma Physics, 17491 Greifswald, GermanyWendelstein 7-X Operations Division, Max Planck Institute for Plasma Physics, 17491 Greifswald, GermanyMachine protection is a core task of real-time image diagnostics aiming for steady-state operation in nuclear fusion devices. The paper evaluates the applicability of the newest low-power NVIDIA Jetson Xavier NX platform for image plasma diagnostics. This embedded NVIDIA Tegra System-on-a-Chip (SoC) integrates a Graphics Processing Unit (GPU) and Central Processing Unit (CPU) on a single chip. The hardware differences and features compared to the previous NVIDIA Jetson TX2 are signified. Implemented algorithms detect thermal events in real-time, utilising the high parallelism provided by the embedded General-Purpose computing on Graphics Processing Units (GPGPU). The performance and accuracy are evaluated on the experimental data from the Wendelstein 7-X (W7-X) stellarator. Strike-line and reflection events are primarily investigated, yet benchmarks for overload hotspots, surface layers and visualisation algorithms are also included. Their detection might allow for automating real-time risk evaluation incorporated in the divertor protection system in W7-X. For the first time, the paper demonstrates the feasibility of complex real-time image processing in nuclear fusion applications on low-power embedded devices. Moreover, GPU-accelerated reference processing pipelines yielding higher accuracy compared to the literature results are proposed, and remarkable performance improvement resulting from the upgrade to the Xavier NX platform is attained.https://www.mdpi.com/1996-1073/15/6/2088graphics processing unitgeneral-purpose computing on graphics processing unitsimage processingplasma diagnosticsembedded system |
spellingShingle | Bartłomiej Jabłoński Dariusz Makowski Piotr Perek Patryk Nowak vel Nowakowski Aleix Puig Sitjes Marcin Jakubowski Yu Gao Axel Winter The W-X Team Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics Energies graphics processing unit general-purpose computing on graphics processing units image processing plasma diagnostics embedded system |
title | Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics |
title_full | Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics |
title_fullStr | Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics |
title_full_unstemmed | Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics |
title_short | Evaluation of NVIDIA Xavier NX Platform for Real-Time Image Processing for Plasma Diagnostics |
title_sort | evaluation of nvidia xavier nx platform for real time image processing for plasma diagnostics |
topic | graphics processing unit general-purpose computing on graphics processing units image processing plasma diagnostics embedded system |
url | https://www.mdpi.com/1996-1073/15/6/2088 |
work_keys_str_mv | AT bartłomiejjabłonski evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT dariuszmakowski evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT piotrperek evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT patryknowakvelnowakowski evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT aleixpuigsitjes evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT marcinjakubowski evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT yugao evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT axelwinter evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics AT thewxteam evaluationofnvidiaxaviernxplatformforrealtimeimageprocessingforplasmadiagnostics |