Digging Deeper in Real-time for Security, Reliability and Performance: Dynamic and Multi-Granular Analytics for Virtualized Networks

Real-time analytics provides the promise to enable the rapid detection and mitigation of security, reliability, and performance issues in the emerging area of virtualized services and networks. These multi-domain Software Defineded Networks (SDN), emerging in the industry, can be dynamically reconfigured in real-time through software, allowing rapid response to changes in service demands, and can be used in conjunction with cloud-based services. However, these networks also introduce new vulnerabilities and threats to security and reliability, as well as new types of performance issues. As the volume, variety, velocity, and veracity of data in these networks is vast, real-time analytics needs to be customized to provide autonomic actions once a potential problem is detected in the network. In this paper, we describe a real-time analytics architecture {with dynamic and varying granularity in time, data, probability, and specifcity for virtualized networks that supports the detection and mitigation of security, reliability and performance issues in real-time.