Federated Learning Applications In Enterprise Network Management

Abstract

The fast expansion of connected devices and the need for improved security and performance provide growing challenges for enterprise network administration. Because of scalability and privacy issues, traditional centralized network management techniques frequently fail to meet these constraints. By facilitating decentralized, cooperative learning across dispersed network entities while maintaining data privacy, federated learning (FL) shows promise as a remedy. In order to improve network scalability, security, and efficiency, this study explores the use of FL in enterprise network administration. We start by conducting a thorough literature assessment of current approaches in cloud computing, software-defined networking (SDN), and network administration, emphasizing both their advantages and disadvantages. Important studies are reviewed, including the management problems in SDN , security measures for SDN control layers , and GENI's federated testbed for new network experiments. A notable deficiency in the incorporation of federated learning in these settings is noted by the review.