Acceleration of an Algorithm Based on the Maximum Likelihood Bolometric Tomography for the Determination of Uncertainties in the Radiation Emission on JET Using Heterogeneous Platforms
Source abstract
Abstract
In recent years, a new tomographic inversion method based on the Maximum Likelihood (ML) approach has been adapted to JET bolometry. Apart from its accuracy and reliability, the key advantage is its ability to provide reliable estimates of the uncertainties in the reconstructions. The original algorithm was implemented and validated using the MATLAB software tool. This work presents the accelerated version of the algorithm implemented using a compatible ITER fast controller platform with the Ubuntu 18.04 or the ITER Codac Core System distributions (6.1.2). The algorithm has been implemented in C++ using the open-source libraries: ArrayFire, ALGLIB, and MATIO. These libraries simplify the management of specific hardware accelerators such as GPUs and increase performance. The speed-up factor obtained is approximately 10 times. The work presents the methodology followed, the results obtained, and the advantages and drawbacks of implementation.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Nuclear Physics and Applications · Atomic and Subatomic Physics Research · Nuclear reactor physics and engineering
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Instrumentation and Applied Acoustic Research Group, Universidad Politécnica de Madrid, 28031 Madrid, Spain
National Institute for Laser, Plasma and Radiation Physics, Magurele, 077125 Bucharest, Romania
Department of Industrial Engineering, University of Rome ‘Tor Vergata’, Via del Politecnico 1, 00133 Rome, Italy
Laboratorio Nacional de Fusión, CIEMAT, Av. Complutense 40, 28040 Madrid, Spain
Consorzio RFX (CNR, ENEA, INFN, Universita’ di Padova, Acciaierie Venete SpA), Corso Stati Uniti 4, 35127 Padova, Italy
Funding
MCIN/AEI/10.13039/501100011033 · Comunidad de Madrid · Euratom Research and Training Programme
References (26)
Show source-provided references
1. Wesson, J. (2011). Tokamaks, Oxford University Press. [4th ed.].
2. Improved radiation measurements on JET–First results from an upgraded bolometer system10.1016/j.jnucmat.2007.01.124 ↗
3. Upgraded bolometer system on JET for improved radiation measurements10.1016/j.fusengdes.2007.03.027 ↗
4. Soft X ray tomography during ELMs and impurity injection in JET10.1088/0029-5515/38/11/307 ↗
5. Current Research into Applications of Tomography for Fusion Diagnostics10.1007/s10894-018-0178-x ↗
6. Maximum likelihood bolometric tomography for the determination of the uncertainties in the radiation emission on JET TOKAMAK10.1063/1.5027880 ↗
7. A comprehensive study of the uncertainties in bolometric tomography on JET using the maximum likelihood method10.1063/1.5119441 ↗
8. On the effects of missing chords and systematic errors on a new tomographic method for JET bolometry10.1016/j.fusengdes.2019.03.120 ↗
9. Investigating the thermal stability of highly radiative discharges on JET with a new tomographic method10.1088/1741-4326/ab7536 ↗
10. Deep learning for plasma tomography using the bolometer system at JET10.1016/j.fusengdes.2016.11.006 ↗
11. Monitoring the plasma radiation profile with real-time bolometer tomography at JET10.1016/j.fusengdes.2020.112179 ↗
12. Fuchs, J., Neu, R., Dux, R., Fuchs, J.C., Junker, W., Kallenbach, A., Mertens, V., de PefiarHempel, S., Schonmann, K., and ASDEX Upgrade Team (July, January 27). Influence of the Carbon and the Oxygen Concentration on the Density Limit in ASDEX Upgrade ECA. Proceedings of the 21st EPS Conference Contribution, Montpellier, France.
13. Description of complex viewing geometries of fusion tomography diagnostics by ray-tracing10.1063/1.5031087 ↗
14. Optimization of the computation of total and local radiated power at ASDEX Upgrade10.1088/1741-4326/abf2e1 ↗
15. Tomographic reconstruction of bolometry for JT-60U diverted tokamak characterization10.1088/0741-3335/43/7/309 ↗
16. Porting Matlab Applications to High-Performance C++ Codes: CPU/GPU-Accelerated Spherical Deconvolution of Diffusion MRI Data10.1007/978-3-319-49583-5_49 ↗
17. Soft real-time EPICS extensions for fast control: A case study applied to a TCV equilibrium algorithm10.1016/j.fusengdes.2014.03.044 ↗
18. Application of Heterogeneous Computing Techniques for the Development of an Image-Based Hot Spot Detection System Using MTCA10.1109/TNS.2021.3087124 ↗
19. Methodology for the implementation of real-time image processing systems using FPGAs and GPUs and their integration in EPICS using Nominal Device10.1016/j.fusengdes.2018.02.051 ↗
20. (2022, May 01). ArrayFire. Available online: https://arrayfire.com/.
21. (2022, May 01). ALGLIB. Available online: https://www.alglib.net/.
22. (2022, May 01). MATIO. Available online: https://github.com/tbeu/matio.
23. A GPU fully vectorized approach to accelerate performance of NSGA-2 based on stochastic non-domination sorting and grid-crowding10.1016/j.asoc.2019.106047 ↗
24. Suitability of recent hardware accelerators (DSPs, FPGAs, and GPUs) for computer vision and image processing algorithms10.1016/j.image.2018.07.007 ↗
25. (2022, May 01). Performance of ArrayFire JIT Code Generation. Available online: https://arrayfire.com/blog/performance-of-arrayfire-jit-code-generation/.
26. A methodology to standardize the development of FPGA-based high-performance DAQ and processing systems using OpenCL10.1016/j.fusengdes.2020.111561 ↗