This article tackles the critical challenge of optimizing the buffer pool, a core component of Database Management Systems (DBMS) that caches frequently accessed data pages, where manual configuration often proves inadequate in dynamic, high-demand environments. To address this gap, we present an automated, data-driven methodology that combines advanced Machine Learning techniques with Bayesian optimization. Our approach follows a systematic three-phase process: (1) Exploratory Factor Analysis (EFA) coupled with K-means clustering to uncover latent factors and reduce the dimensionality of performance metrics; (2) LASSO regression to identify and rank the most influential configuration parameters; and (3) Bayesian optimization using Gaussian Process modeling with acquisition functions (Expected Improvement, Probability of Improvement, and Upper Confidence Bound) to fine-tune buffer pool settings. The main contributions of this work include a novel automated framework for DBMS tuning that simplifies configuration, enhances memory management, and boosts performance efficiency. We validated the proposed solution using real workloads collected from a large-scale financial system in Latin America, achieving up to a 45% reduction in maximum data access wait times, confirming improvements in performance and scalability.

Forest fires have the potential to cause enormous social, economic and, above all, environmental damage. Nowadays, the intelligent and early detection of forest fires is a fundamental technological tool for rescue and emergency agencies in mitigating damage. Among the possibilities, video surveillance technology in conjunction with artificial intelligence and computer vision techniques proves to be an interesting solution. This work proposes the SEMFOGO system, an intelligent system for monitoring and rapid detection of fires in the cerrado region, which currently in South America has an area of 2,045,000 km2between Brazil, Bolivia and Paraguay. The SEMFOGO solution uses the massive capture and processing of video streams through a distributed and scalable system and applies a deep learning model that performs a grid classification on parts of the image. This work also presents a new training dataset specific to the cerrado region with smoke contour annotations. The experimental results compared the metrics of several models and show the practical feasibility of the proposed solution for real time fire detection.

Fifth Generation of Mobile Communication (5G) integrates the use of telecommunication and computer systems. As virtualisation eases the deployment of new functionalities demanded by many industrial and social use cases, also show many research challenges regarding performance and resource optimisation. In the 5G architecture, while mobile networks are already trying to implement a full virtualisation of hardware resources, the core itself lacks of an integrated performance evaluation proposal. In this paper we propose a performance evaluation framework, based on an evaluation function and an assortment of distributed observation functions acting as monitors for compute nodes in a virtualised. infrastructure. The framework, unlike previous proposals, is designed to be integrated into 5G as a native service. The framework was evaluated in a Ultra-Reliable and Low Latency Communications (URLLC) scenario and the results show success in monitoring and analysing a Network Function Virtualisation (NFV) environment with a standard 5G NFV implementation.

Internet of Things (IoT) is one of the key enabling technologies for an always-connected world and also a main enabler for generating information of interest in various application domains. A growing problem in recent years in this technology is security, as power-constrained devices that are typical of IoT applications may not always provide these implementations properly. These conditions can compromise entire environments and allow malicious agents to take control and perform malicious activities. In this article, we provide a summary of the principal vulnerabilities reported for IoT devices based on the OWASP Internet of Things Project, classified by test routine groups. Using models based on standard architectures to define and detail reproducible verification routines for each test, a selection of independent analyzes of each identified category was performed to ensure more comprehensive and accurate testing. Finally, the proposed routines are performed in a test environment to exemplify and ensure their operation, thus contributing to meeting the demand in the area for more accurate information and to assist in understanding the most common vulnerabilities.

The radio spectrum is a finite resource and the current spectrum licensing model has led to a fenomenon known as spectrum scarcity. Several approaches have been presented in order to improve the effectiveness of spectrum utilization. This works main contribution is is to present an approach to opportunistic spectrum sensing and allocation focused on maximizing spectrum usage while minimizing communication time and interference on primary communications. Compared an alternative, our approach presents significant improvements in the spectrum access and utilization, improving network throughput up to 16 times.

The growing demand for mobile wireless access has stimulated the emergence of new communication technologies. Opportunistic Spectrum Access (OSA) is viewed as a promising alternative to overcome the problems caused by static spectrum assignment. Opportunistic access allows dynamic mapping of the transmission needs and communication opportunities. However, performing this task efficiently is not trivial. Indeed, it has been shown to be NP-complete. In this context, this paper presents an efficient heuristic for solving the channel allocation and data assignment problem, according to the opportunities and channels available. The proposed heuristic is optimal in terms of energy consumption, being close to the optimum, about 5% above, in terms of transmission time.