Performance comparison of container orchestration platforms with low cost devices in the fog, assisting Internet of Things applications

Fayos-Jordan, Rafael; Felici-Castell, Santiago; Segura-García, Jaume; Lopez-Ballester, Jesus; Cobos, Máximo

doi:10.1016/j.jnca.2020.102788

Cited by 38 publications

(24 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Additionally, by increasing the number of functions and the number of connected devices to the fog and edge Layers, an additional orchestration layer is required for management purposes. The authors in [38] provide thorough functional and performance comparison with container orchestrators (in particular Docker Swarm and Kubernetes) under different real topologies in the fog-based cluster, using wired and wireless infrastructures. An example of an Edge-AI-based architecture is depicted in Figure 3.…”

Section: Central Architecture (Cloud-based)mentioning

confidence: 99%

GNSS-Free Outdoor Localization Techniques for Resource-Constrained IoT Architectures: A Literature Review

et al. 2021

View full text Add to dashboard Cite

Large-scale deployments of the Internet of Things (IoT) are adopted for performance improvement and cost reduction in several application domains. The four main IoT application domains covered throughout this article are smart cities, smart transportation, smart healthcare, and smart manufacturing. To increase IoT applicability, data generated by the IoT devices need to be time-stamped and spatially contextualized. LPWANs have become an attractive solution for outdoor localization and received significant attention from the research community due to low-power, low-cost, and long-range communication. In addition, its signals can be used for communication and localization simultaneously. There are different proposed localization methods to obtain the IoT relative location. Each category of these proposed methods has pros and cons that make them useful for specific IoT systems. Nevertheless, there are some limitations in proposed localization methods that need to be eliminated to meet the IoT ecosystem needs completely. This has motivated this work and provided the following contributions: (1) definition of the main requirements and limitations of outdoor localization techniques for the IoT ecosystem, (2) description of the most relevant GNSS-free outdoor localization methods with a focus on LPWAN technologies, (3) survey the most relevant methods used within the IoT ecosystem for improving GNSS-free localization accuracy, and (4) discussion covering the open challenges and future directions within the field. Some of the important open issues that have different requirements in different IoT systems include energy consumption, security and privacy, accuracy, and scalability. This paper provides an overview of research works that have been published between 2018 to July 2021 and made available through the Google Scholar database.

show abstract

Section: Central Architecture (Cloud-based)mentioning

confidence: 99%

GNSS-Free Outdoor Localization Techniques for Resource-Constrained IoT Architectures: A Literature Review

et al. 2021

View full text Add to dashboard Cite

show abstract

“…However, as the number of users increases to 500, the response time gap between DS and K8s shrinks, but DS still has overall less response time than K8s has. A similar study targeting fog computing in IoT systems context is presented in [32]. Authors in this study have used different Small Board Computers (SBCs) such as Raspberry Pi 3, Udoo, Tinker, and Jetson to make a fog cluster where one device is master and others are slave.…”

Section: Literature Reviewmentioning

confidence: 99%

Empirical Performance and Energy Consumption Evaluation of Container Solutions on Resource Constrained IoT Gateways

Raza

Jeong

Kim

et al. 2021

Sensors

View full text Add to dashboard Cite

Containers virtually package a piece of software and share the host Operating System (OS) upon deployment. This makes them notably light weight and suitable for dynamic service deployment at the network edge and Internet of Things (IoT) devices for reduced latency and energy consumption. Data collection, computation, and now intelligence is included in variety of IoT devices which have very tight latency and energy consumption conditions. Recent studies satisfy latency condition through containerized services deployment on IoT devices and gateways. They fail to account for the limited energy and computing resources of these devices which limit the scalability and concurrent services deployment. This paper aims to establish guidelines and identify critical factors for containerized services deployment on resource constrained IoT devices. For this purpose, two container orchestration tools (i.e., Docker Swarm and Kubernetes) are tested and compared on a baseline IoT gateways testbed. Experiments use Deep Learning driven data analytics and Intrusion Detection System services, and evaluate the time it takes to prepare and deploy a container (creation time), Central Processing Unit (CPU) utilization for concurrent containers deployment, memory usage under different traffic loads, and energy consumption. The results indicate that container creation time and memory usage are decisive factors for containerized micro service architecture.

show abstract

“…Em busca de uma plataforma de orquestrac ¸ão que possa atender de maneira eficiente a criac ¸ão de clusters para o processamento de dados de aplicac ¸ões IoT, [Fayos-Jordan et al 2020][1] apresentam um estudo baseado na realizac ¸ão de testes de desempenho dos principais orquestradores disponíveis executando em dispositivos com poder computacional limitado (Small Board Computers). Após as análises, os autores concluíram que, embora o Kubernetes apresente um conjunto de recursos que pode ser útil em computadores de alto desempenho, o Swarm provê a utilizac ¸ão mais eficiente dos recursos de processamento em dispositivos com baixo poder computacional.…”

Section: Trabalhos Relacionadosunclassified

“…Diferente dos tradicionais servidores físicos ou das máquinas virtuais, contêineres, de modo geral, podem ser considerados um tipo de virtualizac ¸ão leve. Neles, é possível incluir servic ¸os, dados e códigos de forma isolada do sistema operacional do nó onde estão em execuc ¸ão, reduzindo a utilizac ¸ão de espac ¸o em disco e o consumo dos recursos de processamento [Fayos-Jordan et al 2020, Do Espírito Santo et al 2019, Hoque et al 2017, Docker 2021.…”

Section: Requisitos Da Infraestruturaunclassified

“…Embora possuam objetivos semelhantes, cada orquestrador apresenta seus mecanismos particulares para descoberta de nós, políticas de acesso e gerenciamento de toda a infraestrutura, incluindo o gerenciamento da comunicac ¸ão entre os nós, contêineres e servic ¸os. Isso faz com que cada um tenha suas vantagens e desvantagens, podendo apresentar melhor ou pior performance em determinados cenários [Fayos-Jordan et al 2020].…”

Section: Requisitos Da Infraestruturaunclassified

See 1 more Smart Citation

MSOFEC: Orquestração Multicamada Névoa-Nuvem para Colocação de Serviços em Internet das Coisas Industrial

Contini

Madeira

Bittencourt

2021

Anais Do XXXIX Simpósio Brasileiro De Redes De Computadores E Sistemas Distribuídos (SBRC 2021)

View full text Add to dashboard Cite

Com o surgimento da Quarta Revolução Industrial, diversas fábricas passaram a adotar um modelo de produção que, além de automatizado, opera com a coleta e análise dos dados por meio de diversos sensores e aplicações a ﬁm de melhorar a eﬁciência e produtividade. Com isso, a quantidade de dados coletados e transmitidos aumentou consideravelmente. Nesse contexto, este trabalho propõe um mecanismo de orquestração de serviços e dados baseado na análise periódica das métricas de rede e recursos de processamento de modo que a aplicação obtenha níveis maiores de desempenho e conﬁabilidade na criação da infraestrutura e no direcionamento dos dados em um cenário multicamada névoa-nuvem. Os resultados iniciais demonstraram ser possível diminuir a latência e melhorar a utilização de recursos de processamento, mantendo-se bons níveis de disponibilidade e conﬁabilidade para a aplicação.

show abstract

Performance comparison of container orchestration platforms with low cost devices in the fog, assisting Internet of Things applications

Cited by 38 publications

References 14 publications

GNSS-Free Outdoor Localization Techniques for Resource-Constrained IoT Architectures: A Literature Review

GNSS-Free Outdoor Localization Techniques for Resource-Constrained IoT Architectures: A Literature Review

Empirical Performance and Energy Consumption Evaluation of Container Solutions on Resource Constrained IoT Gateways

MSOFEC: Orquestração Multicamada Névoa-Nuvem para Colocação de Serviços em Internet das Coisas Industrial

Contact Info

Product

Resources

About