On resource consumption of machine learning in communications network security

Abstract

As the complexity of communication networks continues to increase, driven by a diverse array of devices, services and applications, the adoption of Machine Learning (ML) has seen a significant rise to address various challenges ranging from management to security. Regarding network security, the application of ML ranges from preventive measures to detection and remediation due to its ability to dynamically learn and adapt to evolving threat landscapes. However, ML requires a significant amount of resources, mainly due to the fact that ML operates on data, and the volumes of data are consistently rising. This review article explores the resource consumption aspect of ML techniques used for network security and provides a comprehensive review of the current state of research. Moreover, we propose a taxonomy that can be used to classify the methods through which the resource consumption can be reduced for different ML-based network security implementations. The focus of the study encompasses several key aspects related to resource consumption, including energy, computing, memory, latency, bandwidth, and human resources. These resources are critical in improving the efficiency and optimizing the reliability and sustainability of network security solutions. Furthermore, based on an extensive literature review, we summarize key points regarding optimizing resource consumption in ML-based network security solutions. Finally, the challenges and future research directions for resource-efficient, ML-based network security solutions are outlined to aid in the advancement of research in this area.