Energy advancement and innovation have generated several challenges for large modernized cities, such as the increase in energy demand, causing the appearance of the small power grid with a local source of supply, called the Microgrid. A Microgrid operates either connected to the national centralized power grid or singly, as a power island mode. Microgrids address these challenges using sensing technologies and Fog-Cloudcomputing infrastructures for building smart electrical grids. A smart Microgrid can be used to minimize the power demand problem, but this solution needs to be implemented correctly so as not to increase the amount of data being generated. Thus, this paper proposes the use of Fog computing to help control power demand and manage power production by eliminating the high volume of data being passed to the Cloud and decreasing the requests’ response time. The GridLab-d simulator was used to create a Microgrid, where it is possible to exchange information between consumers and generators. Thus, to understand the potential of the Fog in this scenario, a performance evaluation is performed to verify how factors such as residence number, optimization algorithms, appliance shifting, and energy sources may influence the response time and resource usage.

Context: Based on blockchain technologies, decentralized applications (DApps) are on the rise. Studies reveal that these systems are being developed without adherence to specific methodological criteria. Therefore, it is imperative to establish a systematic approach to their construction, allowing developers to focus on their research problems. Problem: The lack of clear guidelines for creating DApps is a challenge, making it difficult for new researchers and developers to enter this field. Research attempting to fill this gap often fails to address specific issues. Solution: This article is based on a systematic literature review with the purpose of creating a framework to support developers in decision-making, making it easier to choose the necessary technologies for DApp development. IS Theory: This study was developed under the aegis of argumentation theory and its fundamental pillars: data, security, and conclusions. Method: The research adopts a prescriptive approach and relies on a systematic literature review to formulate the decision-making framework. Summary of Results: The systematic literature review revealed the absence of a similar framework to what was developed in this work. This framework provides guidance for selecting decentralized platforms according to DApp requirements. Contributions and Impact in the IS area: The use of the developed framework will simplify the architecture design phase of decentralized information systems, as the choice of a decentralized platform for executing smart contracts guides the use of specific technologies.

Context: Smart cities are the future in terms of good resource and people management practices. Several resources can be managed by smart cities, among them urban mobility and energy consumption stand out.Problem: The brightness of public roads consume energy resources, but they are not always necessary, as is the case with roads with no use. Thus, energy resources are wasted. This work aimed to explore the economy of energy resources by observing the flow of use of public roads.Solution: Sensors observe the roads, and the costs of the edges are sent to the fog that applies a graph algorithm to determine the paths that have movement and, thus, generate the reductions in energy consumption via switching off lamps or reducing their frequency.Information System Theory: Based on the General Systems Theory, the general system is mapped into independent systems: fog sensors (fog) and algorithm execution nodes (nodes with higher processing power). A general ecosystem is formulated making the information system based on the information generated by the observation of the sensors.Method: A simulated environment was proposed to obtain the representation of a region of interest. For this region, a graph approach based on Dijkstra was applied to consider the paths with the highest flow of accesses and, thus, propose energy consumption reductions.Results Summary: The results obtained point to a possibility of saving energy resources in the range of 60 to 79% depending on the type of lamps used, and the size of observed region. The results reinforce the need to explore intelligent resources for shared use resources management.Contributions and Impact in the IS area: Among the contributions to the area are: a manageable information system for public lighting. Use of fog for path management and energy matrix management. This work also contributes to proposing new approaches to the proposed problem, such as using the social context (via social networks) to define optimal paths.

RESUMOCloud computing is a computational model in which providers offeron-demand services to clients in a transparent way. In this context,the efficient provisioning of resources is extremely important, sinceit is directly related to Quality of Service (QoS), with the Service LevelAgreement (SLA) compliance and the cost of the service. In thisway, virtualization becomes a crucial factor in resource provisioning.In this paper, performance evaluations were performed consideringenvironments configured with different virtualization strategies: byvirtual machines and by containers. In the first set of experiments,we evaluated the boot time of the system with the KVM, Xen andDocker tools and with variations in the number of instances andvirtual cores. In the second, the same mechanisms were analyzedduring the execution of different benchmarks (Apache and Smallpt)and with variations in the number of virtual resources provisioned.In the results, Docker proved to be more efficient when comparedto other tools.

As quedas em altura representam um grave problema de saúde pública no Brasil, extrapolando os limites do trabalho formal e exigindo estratégias baseadas em evidências. Este estudo caracterizou perfis de vulnerabilidade em óbitos por quedas de escada de mão (W11), andaimes (W12) e edifícios (W13), utilizando dados do DATASUS (1996-2023). Aplicou-se a mineração de dados (K-Means) para segmentação por sexo, idade e escolaridade, validada pelos métodos cotovelo e silhueta. Os resultados revelaram um núcleo de vulnerabilidade crítica composto predominantemente por homens (50-59 anos) em todas as categorias, destacando-se idosos (60-69 anos) nas quedas de escada de mão. Nos grupos de maior risco, predominou a escolaridade de 4 a 7 anos(fundamental incompleto). Conclui-se que o K-Means isolou eficazmente padrões de risco, evidenciando que a fatalidade atinge severamente trabalhadores em envelhecimento e baixa instrução, em contextos laborais e informais.

Computação em nuvem é um estilo de computação no qual os provedores de recursos podem oferecer serviços sob demanda de forma transparente e os clientes geralmente pagam de acordo com o uso. A nuvem introduz um novo nível de flexibilidade e escalabilidade para usuários abordando desafios como a rápida alteração em cenários de Tecnologia de Informação (TI) e a necessidade de reduzir custos e tempo no gerenciamento de infraestrutura. No entanto, para ser capaz de oferecer garantias de qualidade de serviço (QoS) sem limitar o número de requisições aceitas, os provedores devem ser capazes de escalonar de forma dinâmica e eficiente as requisições de serviços para serem executadas nos recursos disponíveis. O balanceamento de carga não é uma tarefa trivial, envolvendo desafios relacionados à demanda de serviço, a qual pode mudar instantaneamente, à modelagem de desempenho, e implantação e monitoramento de aplicações em recursos de TI virtualizados. Dessa forma, o objetivo deste artigo é desenvolver e avaliar algoritmos de balanceamento de carga para um ambiente em nuvem de forma a estabelecer um mapeamento mais eficiente entre as requisições de serviços e as máquinas virtuais que as executarão, garantindo a qualidade do serviço conforme definido no acordo de níveis de serviço. Nos experimentos verificou-se que o algoritmo proposto permitiu uma redução no tempo de execução das requisições, aumentando a quantidade de requisições atendidas durante o tempo de observação.