Desarrollo de algoritmos para un Grid de sistemas empotrados
Loading...
Official URL
Full text at PDC
Publication date
2018
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
La programación paralela es indiscutiblemente el futuro próximo de las empresas ya que cada día surgen nuevos problemas que deben ser resueltos lo antes posible, esto se puede ver reflejado en la cantidad de programas que trabajan de forma ineficiente (por la cantidad de tiempo que requieren). Es por ello que se necesitaran de proyectos de investigación que intenten solucionar este tipo de problemas para poder hacer un mejor uso de los recursos.
El grid desarrollado funciona en esta línea, ayuda a realizar un estudio sobre programas ejecutados en paralelo (utilizando MPI) sobre una placa (arquitectura Intel) desarrollada para el IoT (Internet of Things) proporcionando estudios estadísticos sobre si es mejor tener un sistema centralizado o más distribuidos.
Todo esto se consiguió gracias a la implementación de diferentes algoritmos (siendo estos ejecutados de forma secuencial y paralelo) los cuales nos permitieron explotar los diferentes recursos hardware en la placa Intel Galileo Gen 2.
Parallel Programming is undoubtedly the future of companies, with new problems arising every day, is necessary to be solved as soon as possible. This can be seen in the number of programs that work inefficiently (for the amounts of time they require in perform its calculations). That’s why projects like this will be needed to help solve these kinds of problems to make a better use of the resources. The grids developed works in this line, it helps to carry out a study on programs executed in parallel (using MPI) on a board (Intel architecture) developed for the IoT (Internet of Things) providing statistical studies on whether it is better to have a system centralized or more distributed. All this was achieved thanks to the implementation of different algorithms (implemented in sequential and parallel) which allowed us to exploit all the hardware resources on the board Intel Galileo Gen 2.
Parallel Programming is undoubtedly the future of companies, with new problems arising every day, is necessary to be solved as soon as possible. This can be seen in the number of programs that work inefficiently (for the amounts of time they require in perform its calculations). That’s why projects like this will be needed to help solve these kinds of problems to make a better use of the resources. The grids developed works in this line, it helps to carry out a study on programs executed in parallel (using MPI) on a board (Intel architecture) developed for the IoT (Internet of Things) providing statistical studies on whether it is better to have a system centralized or more distributed. All this was achieved thanks to the implementation of different algorithms (implemented in sequential and parallel) which allowed us to exploit all the hardware resources on the board Intel Galileo Gen 2.
Description
Universidad Complutense, Facultad de Informática, curso 2016/2017