Online Density Grid Pattern Analysis to Classify Anomalies in Cloud and NFV Systems

摘要: Technologies like machine-to-machine communication, autonomous driving or virtual reality applications form an increasingly diverse service landscape. This entails individual and dynamic requirements regarding scalability, availability, latency throughput from the underlying IT infrastructure. To meet those, telecommunication network providers started a transformation process towards virtualized technologies function virtualization (NFV). However, this drastically increases infrastructure complexity to point where more management is required. In order reliability of dedicated hardware, solutions are in demand recovery remediation systems. For critical systems, actions must be selected very cautiously not disrupt operational process. enable precise handling, anomaly situations need accurately identified based on monitoring data streams. Therefore, we present supervised machine learning method for online classification states similarities between type-specific density grid patterns. evaluation, created extensive NFV testbed running implementation IP multimedia subsystem. Applying our classify various synthetically injected situations, results reveal average overall accuracy 0.94. Further also show that model applicable identifying previously unknown situations. Thus, approach provides valuable step maintenance infrastructures.