Delivering resiliency, security and trust in evolving AN

The key to resiliency in autonomous networks (AN), according to Arvinder Khanna, APAC Cloud and Network Services Pre-sales Leader, Nokia, is understanding their complexity and evolving parameters. For example, networks might be converged and many are likely to be multi-layered across multiple domains.

Next it important to understand traffic patterns, especially with new services in the offing. Third, resiliency is no longer primarily about redundancy, but the ability to respond rapidly to meet SLAs.

Fourth, data has to be high quality, correlated from end-to-end. Khanna noted that sometimes operators run predictive trials with simulated data which provides poor insights and erodes trust. Khanna went on to explain how Nokia addresses these crucial issues for Level 3 and 4 AN.

Security is also critical and again, there are new potential risks: cloud-native architectures and closed loop AN expose the network through APIs meaning that zero trust behaviors should mean exactly that – everything must be safeguarded.

Next, operators should always be roll back work, at every layer, including CI/CD. Finally, it is essential to protect the AI model which is built on top of the data: hackers look to corrupt models and control their outcomes.