With expansion and highly distributed nature of access devices, and the increase in demands for mobility in Cloud-based environs, the network edge has diminished as a demarcation for security purposes. Embedded network surveillance is increasingly migrating to the edge and endpoints.

To push monitoring and surveillance intelligence to where the threat if fostered, means establishing machine learning and behavioral analytics that does not consume precious processing power on connected devices. The flows of access and transaction patterns of the network itself can be leveraged to represent the normal behavioral baselines, and modelling based on threat heuristics can illuminate anomalies. Keeping this distributed embedded intelligence lightweight and with minimal footprint on the networked nodes.