Tecnologías de la Información y las Comunicaciones mediante IoT para la solución de problemas, Procesamiento de Imágenes en Arquitecturas Multiprocesador y estrategias de Tolerancia a Fallos en HPC
Cargando...
Fecha
2023-11-21
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
El proyecto se desarrolló sobre dos pilares fundamentales, por un lado, se explotaron las tecnologías de internet de las cosas (IoT) para la solución de problemáticas actuales, tales como la eficiencia energética, el control de parámetros en cultivos bajo cubierta y el diseño y desarrollo de laboratorios remotos para potenciar la enseñanza a distancia. Adicionalmente, se continuó con la línea anterior sobre la determinación de eficiencia en el procesamiento de imágenes en sistemas reconfigurables y la tolerancia a fallo en sistemas de cómputo de altas prestaciones (HPC).
Se desarrolló un sistema IoT de control de energía escalable que permite minimizar el consumo en diferentes ambientes, en conjunto con la implementación de un sistema informático que provee información precisa de magnitudes climatológicas, del estado del suelo y del nivel de iluminación en invernaderos. Luego de la implementación se obtuvieron resultados muy satisfactorios respecto a la eficiencia energética, al lograr independizar el consumo del factor humano. Además, se implementaron interfaces wifi y GPRS para el monitoreo y control de parámetros desde plataformas IoT.
Respecto al procesamiento de imágenes en sistemas reconfigurables, se determinó la eficiencia respecto a la ejecución secuencial, mediante la implementación de un sistema multicore (computo paralelo), y la combinación de este último con la implementación concurrente del algoritmo, en donde se obtuvieron muy buenos resultados. Por último, se diseñó una metodología que permite detectar fallos en arquitecturas multicore, dando buenos resultados en diferentes pruebas.
-
The project was developed on two fundamental pillars, on the one hand, Internet of Things (IoT) technologies were exploited to solve current problems, such as energy efficiency, control of parameters in crops under cover and design and development remote laboratories to enhance distance learning. Additionally, the previous line on the determination of efficiency in image processing over reconfigurable systems and falult tolerance in high-performance computing (HPC) systems was continued. A scalable energy control IoT system was developed that allows to minimize consumption in different environments, in conjunction with the implementation of a computer system that provides precise information on climatological magnitudes, the state of the soil and the level of illumination in greenhouse. After the implementation, very satisfactory results were obtained regarding energy efficiency, by keeping independence of consumption from the human factor. In addition, Wi-Fi and GPRS interfaces were implemented for the monitoring and control of parameters from IoT platforms. Regarding the processing of images in reconfigurable systems, the efficiency was determined with respect to the sequential execution, through the implementation of a multicore system (parallel computing), and the combination of the latter with the concurrent implementation of the algorithm, where very good results were obtained. Finally, a methodology was designed that allows detecting failures in multicore architectures, giving good results in different tests.
The project was developed on two fundamental pillars, on the one hand, Internet of Things (IoT) technologies were exploited to solve current problems, such as energy efficiency, control of parameters in crops under cover and design and development remote laboratories to enhance distance learning. Additionally, the previous line on the determination of efficiency in image processing over reconfigurable systems and falult tolerance in high-performance computing (HPC) systems was continued. A scalable energy control IoT system was developed that allows to minimize consumption in different environments, in conjunction with the implementation of a computer system that provides precise information on climatological magnitudes, the state of the soil and the level of illumination in greenhouse. After the implementation, very satisfactory results were obtained regarding energy efficiency, by keeping independence of consumption from the human factor. In addition, Wi-Fi and GPRS interfaces were implemented for the monitoring and control of parameters from IoT platforms. Regarding the processing of images in reconfigurable systems, the efficiency was determined with respect to the sequential execution, through the implementation of a multicore system (parallel computing), and the combination of the latter with the concurrent implementation of the algorithm, where very good results were obtained. Finally, a methodology was designed that allows detecting failures in multicore architectures, giving good results in different tests.
Descripción
Palabras clave
Internet de las cosas, Procesamiento de Imágenes, Sistemas multiprocesador, Tolerancia a fallos, Cómputo de altas prestaciones, Internet of things, Image Processing, Multicore Systems, Fault Tolerance, High Performance Computing
Citación
Osio, J. R., Montezanti, D. M., Cappelletti, M. Á. y Salvatore, J. E. (2023). Tecnologías de la Información y las Comunicaciones mediante IoT para la solución de problemas, Procesamiento de Imágenes en Arquitecturas Multiprocesador y estrategias de Tolerancia a Fallos en HPC. Contribuciones de Ciencia y Tecnología, 1(1). https://rid.unaj.edu.ar/handle/123456789/2836