arXiv subjects

Jéferson Campos Nobre

Publications and source records attributed to Jéferson Campos Nobre.

4 recordsLinked to original sources

arXiv2026

Dynamic Sampling for Telemetry in Microservices: A Reinforcement Learning and Entropy-Based Approach

Microservices architectures are increasingly deployed in cloud-based distributed environments, making application development and maintenance more dynamic, but also increasing the complexity of troubleshooting and observability. Distributed tracing tools are therefore essential for request analysis and debugging, despite introducing additional overhead that can be amplified by excessive and inefficient data collection. This article proposes RADAR (Reinforcement Learning Agent for Dynamic And Relevant trace sampling), an agent that combines reinforcement learning with a data entropy assessment to achieve more efficient capture of traces relevant to system monitoring, based on the OpenTelemetry standard. RADAR tests different sampling rules to discover which combination is most efficient. A test environment simulating a minimalist online store with several microservices distributed across a Kubernetes cluster served as the basis for the experiments, which evaluated the agent's convergence and the system's performance in terms of resource consumption and collected data quality. Results showed that RADAR reduced network bandwidth consumption by 97.4% and CPU usage by 99.0% compared to full data collection, also outperforming a fixed-rate sampling baseline. Beyond these resource savings, the approach preserved observability of critical scenarios, retaining approximately 85.6% of rare trace patterns and increasing the average entropy of the stored information by approximately 25%, validating the feasibility of using entropy to orchestrate telemetry autonomously and efficiently.

cs.NI↗

arXiv2026

On the Impact of Entropy-based Features

Network anomaly detection is increasingly challenging due to the growing diversity and variability of traffic patterns, which are not always well captured by traditional statistical features. In this work, we explore the use of entropy as an additional feature to support supervised network traffic classification. The main idea is to use entropy to represent variability in selected traffic attributes, complementing conventional descriptors rather than replacing them. We integrate the entropy-based feature into a standard machine learning pipeline and evaluate its impact through a direct comparison between models trained with and without this feature. Experiments conducted on a public intrusion detection dataset show consistent improvements in classification performance, while the additional computational cost remains low. The analysis of confusion matrices indicates a reduction in misclassifications, especially in traffic scenarios with higher variability. Overall, the results suggest that entropy-based features offer a simple and practical way to enhance existing anomaly detection pipelines. This approach is particularly attractive in settings where lightweight feature engineering and interpretability are important, making entropy a useful complement to commonly used traffic features.

cs.CR↗

arXiv2021

On the Privacy of National Contact Tracing COVID-19 Applications: The Coronavírus-SUS Case

The 2019 Coronavirus disease (COVID-19) pandemic, caused by a quick dissemination of the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), has had a deep impact worldwide, both in terms of the loss of human life and the economic and social disruption. The use of digital technologies has been seen as an important effort to combat the pandemic and one of such technologies is contact tracing applications. These applications were successfully employed to face other infectious diseases, thus they have been used during the current pandemic. However, the use of contact tracing poses several privacy concerns since it is necessary to store and process data which can lead to the user/device identification as well as location and behavior tracking. These concerns are even more relevant when considering nationwide implementations since they can lead to mass surveillance by authoritarian governments. Despite the restrictions imposed by data protection laws from several countries, there are still doubts on the preservation of the privacy of the users. In this article, we analyze the privacy features in national contact tracing COVID-19 applications considering their intrinsic characteristics. As a case study, we discuss in more depth the Brazilian COVID-19 application Coronavírus-SUS, since Brazil is one of the most impacted countries by the current pandemic. Finally, as we believe contact tracing will continue to be employed as part of the strategy for the current and potential future pandemics, we present key research challenges.

cs.CR↗

arXiv2016

Consistency Maintenance of State of Management Data in P2P-based Autonomic Network Management

Complex Dynamic Networks can be exploited in solving problems where traditional solutions may not be sufficient. The increasing complexity of computer networks imposes problems to the current network management solutions. In this context, network management is an example of a research area that could benefit from the use of CDNs. However, the consistency of state of management data among the elements that build management CDNs (management nodes) is an important challenge. Traditional mechanisms to maintain consistency of these states are supported by some centralization which wastes some desirable properties of CDNs (e.g., robustness). In contrast to these mechanisms, we propose a distributed, scalable and robust mechanism to maintain the consistency of state of management data in management CDNs. Our mechanism introduces multi-agent truth maintenance features and communication strategies based on dynamic process to provide consistency maintenance of state of management data. We developed a model of a management CDN on Peersim simulator to perform experiments. Besides, 2 case studies are presented. The result obtained supports our scalability and robustness claims.

cs.NI↗