‘Breakthrough’ Cisco New Network Uses Machine Learning To Detect Malware In Encrypted Data

Cisco says its ‘new network’ is one of the most “significant breakthroughs” ever made in enterprise network, using machine learning to adapt to business needs and detect threats in encrypted traffic without impacting privacy.

The key distinguishing features of this new network platform are automation, visibility and software, accelerating the shift from hardware which has taken place over the past few years. Cisco itself has expanded from its roots into security and services.

All of this, the company claims, is essential as mobility, cloud and the Internet of Things (IoT) will make it increasingly complex to manage network policies and the number of connected devices.

The new network

“Today we are unveiling a new network for a new era, a network for today’s world and tomorrow’s. Our new network is the result of the deep innovation we’ve been driving at Cisco, and it’s something I believe will change the trajectory of the entire industry,” said Cisco CEO Chuck Robbins.

“The new network delivers a world where you can connect billions of devices, identify them almost instantly, know what’s trustworthy and what isn’t, and draw exponential value from the connections – and you can do it in hours instead of weeks and months. This capability is so new and so vital, that in our view, it will free up businesses to pursue new opportunities — because big changes will seem less daunting and less risky.

“We are creating an intelligent platform with unmatched security for today and for the future that propels businesses forward and creates new opportunities for people and organizations everywhere.”

Cisco claims that the majority of the world’s web traffic passes through its equipment, giving it a vast treasure trove of data to analyse for threats.

Of course, much of this is encrypted, but by combining the metadata of the communications with intelligence from Cisco’s Talos cybersecurity division, the Encrypted Traffic Analytics (ETA) can recognise and mitigate threats with 99 percent accuracy – without decrypting traffic.

Encryption challenge

“Cisco’s [ETA] solves a network security challenge previously thought to be unsolvable,” added David Goeckeler, general manager of networking and security at Cisco. “ETA uses Cisco’s Talos cyber intelligence to detect known attack signatures even in encrypted traffic, helping to ensure security while maintaining privacy.”

Machine learning is also applied through the Network Data Platform. This categorises data running in the network and turns it into predictions an actionable insights, giving organisations more information about their business and allowing admins to make changes.

Software-Defined Access allows for automated policy enforcement and network segmentation  – automatically performing tasks like configuration, provisioning and troubleshooting. When combined with a new centralised dashboard called the ‘DNA Center’, IT departments have greater visibility and can save time on performing common tasks.

“Initial analysis with field trial customers and internal testing have shown a reduction in network provisioning time by 67 percent, improved issue resolution by 80 percent [and] reduced security breach impact by 48 percent,” Cisco claims.

The new network platform will be supported by new Catalyst 9000 switches, software subscriptions and other services when it arrives later this year. The new network is being tested by more than 75 customers including NASA and Royal Caribbean cruises.

The company has been talking about the complexities of IoT and growing cybersecurity threats for a number of years now, launching a Threat Defence Service a few months ago. Most recently it agreed an enterprise security partnership with IBM.

Quiz: What do you know about the Internet of Things?