Fault tolerance of network routers using machine learning techniques

Companies are adopting digital technologies at an exponential rate to increase their productivity. To remain competitive, they need to integrate the latest digital innovations into their processes and IT systems with the latest digital innovations such as mobility, data analytics, the cloud, and the Internet of Things (IoT). Network deployment is a matter of seconds through automation. This has created new configuration or design issues. To solve these routing problems, fault tolerance is used, to provide high timeliness and accuracy in a system. This chapter studies fault tolerance problems in a distributed network system, including large-scale networks such as wide area networks (WANs) and metropolitan area networks (MANs). We propose an approach based on machine learning algorithms such as K-nearest neighbor (KNN) and support vector machine (SVM) to solve the fault tolerance problem in scheduled jobs. We used the KNN algorithm to identify the generation of a reset of failed cluster jobs by calculating the energy radius and distance of the jobs. The SVM is used to classify the fault tolerance jobs by considering the hybrid, core, and single node to analyze the failure of the job at different levels with the respective nodes. The results prove the effectiveness of the proposed approach for fault tolerance analysis of network routers.