Abstract
Image-based diagnostics are key for fusion experiments. The operating conditions at ITER and the future machines require changing the role of such systems from monitoring and archiving for offline postprocessing to real-time processing. One of the roles of such systems is machine protection. A relevant application of vision diagnostics is the wall and divertor temperature monitoring and hot spot detection. However, algorithms for hot spot detection are computationally costly. To achieve real-time performance at the required time resolution for all these experiments, evaluating and validating the newest technologies is vital. This work applies heterogeneous computing techniques based on the OpenCL standard to the real-time hot spot detection problem and obtains the performance values in a Micro Telecommunications Computer Architecture (MTCA) platform. OpenCL reduces the development time, improves portability, and simplifies the evaluation and validation of each part of the algorithm to find the best-suited device in the heterogeneous system. The proposed solution enables balancing the computational load between a field-programmable gate array (FPGA) and a graphical processing unit (GPU). The algorithm has been adapted and optimized, taking profit on the particularities of each platform.
Highlights
T HE use of advanced data acquisition (DAQ) and processing systems in big science experiments is essential to their control and to study the properties and behavior of complex physical phenomena
The following tables show the connected components labeling (CCL) results obtained for graphical processing unit (GPU), for images with 20% of foreground pixels for different image sizes using the original union-find and union-find modified with the foreground pixel list, where Cycles/Px means the number of clock cycles needed to process a pixel and Px/s is the number of pixels processed in a second
Regarding the use of OpenCL for the development of heterogeneous systems, we can state that the OpenCL programming framework is a useful resource to integrate the different parts of a heterogeneous system because the developer can use the same code on different hardware platforms, which is helpful during all the development cycle and simplifies the distribution of the algorithms among the hardware platforms
Summary
T HE use of advanced data acquisition (DAQ) and processing systems in big science experiments is essential to their control and to study the properties and behavior of complex physical phenomena. One example of technology solving this is the OpenCL standard [8] This allows using the same programming language for CPUs, GPUs, and FPGAs. One of the common functions of image DAQ and processing systems in fusion devices is hot spot detection based on visible or infrared image processing [4], [6], [9], [10]. The main goal of this article is to present an example of how OpenCL can help to prototype and evaluate the implementation of the connected components labeling (CCL) algorithm for the hot spot detection algorithm in fusion devices using an MTCA platform and what algorithm’s steps are more suitable for the FPGA or the GPU to get the best performance.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
