| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 2.71 MB | Adobe PDF |
Advisor(s)
Abstract(s)
No mundo da Internet of Things (IoT) surgiram soluções tecnológicas de sistemas
para armazenar e processar na Cloud grandes volumes de dados originados em IoT.
Os dispositivos IoT utilizam comunicações sem fios criando a necessidade de níveis
intermédios (fog computing) para agregar e filtrar os dados antes de serem enviados
para a Cloud. Para além de gerar dados, por exemplo, valores de temperatura,
de sinais vitais como o ritmo cardíaco, alguns dispositivos podem também executar
ações.
Contrariamente ao fluxo de dados Dispositivos-Cloud, é muito importante considerar
cenários Cloud-Dispositivos onde os dados são enviados da Cloud para os
dispositivos, que podem ser utilizados por pessoas, durante a execução de qualquer
atividade. Consequentemente identifica-se um problema - como coordenar as
ações que cada pessoa deve executar, ao receber dados no seu dispositivo IoT.
Esta tese propõe como solução o sistema Rain Computing como analogia com a
chuva que cai das nuvens e inunda rios e lagos. Assim, dados previamente definidos
na Cloud (planos de ação) são enviados para inundar dispositivos, com intermediação
de coordenadores, que sincronizam as ações executadas em cada dispositivo.
A arquitetura do sistema Rain Computing assenta em módulos desacoplados, com
interfaces e mensagens de comunicação bem definidas: i) O Centro de Controlo
executa-se na Cloud e permite aos utilizadores definir e agendar a execução de
planos de ação em múltiplos dispositivos; ii) O módulo intermédio Coordenador
executa-se num servidor ao nível fog computing e sincroniza a execução das ações
do plano nos dispositivos; iii) Adicionalmente também é disponibiliza uma biblioteca
de suporte ao desenvolvimento de aplicações em múltiplos cenários.
São também apresentados dois estudos de caso, o Maestro Digital e o Luzes de
Palco que permitiram avaliar e validar a operacionalidade do sistema Rain Computing
para o desenvolvimento de múltiplos cenários de controlo de dispositivos
IoT.
In the world of Internet of Things (IoT), technological solutions naturally emerged with systems based on large volumes of data originating from IoT devices and processed in the Cloud. IoT devices use wireless communications creating the need for intermediate levels (fog computing) to aggregated and filtered before being sent to the Cloud. In addition to generating data, for example, temperature values, vital signs such as heart rate, some devices can also perform actions. Contrary to the data flow Devices-Cloud, it is very important to consider scenarios Cloud-Devices where data is sent from Cloud to devices, which can be used by people, during the execution of any activity, consequently a problem is identified - how to coordinate the actions that each person must perform, when receiving data on his IoT device. This thesis proposes as a solution the Rain Computing system as an analogy with the rain that falls from the clouds and floods rivers and lakes. Thus, data previously defined in the Cloud (action plans) are sent to flood devices, with the intermediation of coordinators, who synchronize the actions performed on each device.. The Rain Computing system architecture is based on decoupled modules, with welldefined communication messages and interfaces: i) The Control Center module runs on Cloud and allows users to define and schedule the execution of action plans on multiple devices; ii) The intermediate module Coordinator runs on a server at the fog computing level and synchronizes the execution of the plan’s actions on the devices; iii) In addition, a library to support application development in multiple scenarios is also available. Two case studies are presented, Maestro Digital and Luzes de Palco, which allowed to evaluate and validate the operability of the Rain Computing system for the development of multiple control scenarios for IoT devices.
In the world of Internet of Things (IoT), technological solutions naturally emerged with systems based on large volumes of data originating from IoT devices and processed in the Cloud. IoT devices use wireless communications creating the need for intermediate levels (fog computing) to aggregated and filtered before being sent to the Cloud. In addition to generating data, for example, temperature values, vital signs such as heart rate, some devices can also perform actions. Contrary to the data flow Devices-Cloud, it is very important to consider scenarios Cloud-Devices where data is sent from Cloud to devices, which can be used by people, during the execution of any activity, consequently a problem is identified - how to coordinate the actions that each person must perform, when receiving data on his IoT device. This thesis proposes as a solution the Rain Computing system as an analogy with the rain that falls from the clouds and floods rivers and lakes. Thus, data previously defined in the Cloud (action plans) are sent to flood devices, with the intermediation of coordinators, who synchronize the actions performed on each device.. The Rain Computing system architecture is based on decoupled modules, with welldefined communication messages and interfaces: i) The Control Center module runs on Cloud and allows users to define and schedule the execution of action plans on multiple devices; ii) The intermediate module Coordinator runs on a server at the fog computing level and synchronizes the execution of the plan’s actions on the devices; iii) In addition, a library to support application development in multiple scenarios is also available. Two case studies are presented, Maestro Digital and Luzes de Palco, which allowed to evaluate and validate the operability of the Rain Computing system for the development of multiple control scenarios for IoT devices.
Description
Projecto final para obtenção do grau de mestre em Engenharia Informática e de Computadores
Keywords
Cloud computing Fog computing Internet of Things Coordenação Distribuída Distributed Coordination
Citation
SANTOS, António Peixoto de Assunção - Rain Computing : orquestração de ações executadas por grupos de dispositivos IoT. Lisboa: Instituto Superior de Engenharia de Lisboa, 2020. Dissertação de mestrado.
Publisher
Instituto Superior de Engenharia de Lisboa