A Machine-Learning-Based Cyberattack Detector for a Cloud-Based SDN Controller

Mozó, Alberto; Karamchandani, Amit; Cal, Luis de la; Gómez-Canaval, Sandra; Pastor, Antonio; Gifre, Lluís

doi:10.3390/app13084914

Cited by 5 publications

(1 citation statement)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the realm of cloud-based cyber threat intelligence, ref. [22] presents a machine learning-based cyberattack detector for a Cloud-Based SDN Controller. The study integrates robust machine learning components into the TeraFlowSDN (TFS) controller to safeguard against potential malicious actors.…”

Section: Cyber Threat Intelligence In Cloud Environmentsmentioning

confidence: 99%

Evaluation of the Omni-Secure Firewall System in a Private Cloud Environment

Mahmood,

Hasan,

Yahaya

et al. 2024

Knowledge

View full text Add to dashboard Cite

This research explores the optimization of firewall systems within private cloud environments, specifically focusing on a 30-day evaluation of the Omni-Secure Firewall. Employing a multi-metric approach, the study introduces an innovative effectiveness metric (E) that amalgamates precision, recall, and redundancy considerations. The evaluation spans various machine learning models, including random forest, support vector machines, neural networks, k-nearest neighbors, decision tree, stochastic gradient descent, naive Bayes, logistic regression, gradient boosting, and AdaBoost. Benchmarking against service level agreement (SLA) metrics showcases the Omni-Secure Firewall’s commendable performance in meeting predefined targets. Noteworthy metrics include acceptable availability, target response time, efficient incident resolution, robust event detection, a low false-positive rate, and zero data-loss incidents, enhancing the system’s reliability and security, as well as user satisfaction. Performance metrics such as prediction latency, CPU usage, and memory consumption further highlight the system’s functionality, efficiency, and scalability within private cloud environments. The introduction of the effectiveness metric (E) provides a holistic assessment based on organizational priorities, considering precision, recall, F1 score, throughput, mitigation time, rule latency, and redundancy. Evaluation across machine learning models reveals variations, with random forest and support vector machines exhibiting notably high accuracy and balanced precision and recall. In conclusion, while the Omni-Secure Firewall System demonstrates potential, inconsistencies across machine learning models underscore the need for optimization. The dynamic nature of private cloud environments necessitates continuous monitoring and adjustment of security systems to fully realize benefits while safeguarding sensitive data and applications. The significance of this study lies in providing insights into optimizing firewall systems for private cloud environments, offering a framework for holistic security assessment and emphasizing the need for robust, reliable firewall systems in the dynamic landscape of private clouds. Study limitations, including the need for real-world validation and exploration of advanced machine learning models, set the stage for future research directions.

show abstract

Section: Cyber Threat Intelligence In Cloud Environmentsmentioning

confidence: 99%

Evaluation of the Omni-Secure Firewall System in a Private Cloud Environment

Mahmood,

Hasan,

Yahaya

et al. 2024

Knowledge

View full text Add to dashboard Cite

show abstract

A methodological framework for optimizing the energy consumption of deep neural networks: a case study of a cyber threat detector

Karamchandani,

Mozo,

Gómez-Canaval

et al. 2024

Neural Comput & Applic

View full text Add to dashboard Cite

The growing prevalence of deep neural networks (DNNs) across various fields raises concerns about their increasing energy consumption, especially in large data center applications. Identifying the best combination of optimization techniques to achieve maximum energy efficiency while maintaining system performance is challenging due to the vast number of techniques available, their complex interplay, and the rigorous evaluation required to assess their impact on the model. To address this gap, we propose an open-source methodological framework for the systematic study of the influence of various optimization techniques on diverse tasks and datasets. The goal is to automate experimentation, addressing common pitfalls and inefficiencies of trial and error, saving time, and allowing fair and reliable comparisons. The methodology includes model training, automatic application of optimizations, export of the model to a production-ready format, and pre- and post-optimization energy consumption and performance evaluation at inference time using various batch sizes. As a novelty, the framework provides pre-configured "optimization strategies" for combining state-of-the-art optimization techniques that can be systematically evaluated to determine the most effective strategy based on real-time energy consumption and performance feedback throughout the model life cycle. As an additional novelty, "optimization profiles" allow the selection of the optimal strategy for a specific application, considering user preferences regarding the trade-off between energy efficiency and performance. Validated through an empirical study on a DNN-based cyber threat detector, the framework demonstrates up to 82% reduction in energy consumption during inference with minimal accuracy loss.

show abstract