Self-driving cars – also known as driverless, autonomous or robotic cars – are quickly becoming reality. Indeed, Google’s self-driving cars are making waves in the industry, with its artificial intelligence (AI) making it possible for cars to interpret how humans drive.
While self-driving cars still have some way to go, AI is already having a significant impact on the way IT runs enterprises. Businesses are making the transition from being automated to autonomous, where machine learning and AI make it possible to create a ‘self-driving’ wide area network (WAN).
Moving from automated to autonomous
Enterprises are already turning to software-defined WAN (SD-WAN) solutions to connect employees consistently and securely to applications, whether the applications are in the data centre or the cloud.
Automation plays a key role in these current SD-WAN offerings, eliminating many of the repetitive and mundane manual steps required to configure and connect remote offices and branch locations. However, automation also has its limitations.
Indeed, automation is not sufficient to translate high-level business goals or intent into specific actions across the network. What’s more, it is not good at dealing with the many unanticipated situations across production WAN deployments. These are areas where machine learning and AI can come into play.
What will a self-driving WAN look like?
When considering the idea of a self-driving WAN, it’s instructive to look at what’s happening with self-driving cars. Each self-driving car has hundreds of sensors that collect information to build a real-time model of the environment surrounding the car. AI is applied to determine how the car should react at any given moment. A combination of classic control loops and newer machine learning algorithms work in tandem to achieve a high-level goal: driving the car safely from point A to point B.
Furthermore, most implementations supplement the car-level intelligence with fleet-level learning. Every car provides data into a central repository where learning across all the vehicles in the fleet is aggregated and applied.
Using data from fleet-learning brings important advantages. One is building more complete and accurate maps, while the other is better identifying hazards and reducing false positives. And, perhaps most importantly, fleet-learning provides a way to track and improve a vehicles’ software performance.
When it comes to the self-driving WAN, it will encompass hierarchical learning in the same way. Indeed, learning will occur at the network-device level, the enterprise level, and – for the enterprises that opt-in – learning will occur in aggregate across a “fleet” of many enterprises.
To take the self-driving car analogy one step further, it’s interesting to look at what Google has done with its Waymo cars. Google removed all the manual controls with the exception of an emergency stop button.
At the insistence of the California Department of Motor Vehicles, Google reportedly reinstated a steering wheel, but otherwise, the primary interface for a self-driving car was high level goal or intent based: safely transporting occupants from point A to point B via the most efficient route.
As the enterprise moves towards the self-driving WAN, the same kind of transition can happen. Instead of having to understand the numerous protocols and manual command-line interface (CLI) controls applied device-by-device, the WAN will be driven by high-level business intent.
As such, the network administrator will be able to focus more on the services the network is intended to deliver, and their impact on the business, and less on the underlying details of how that happens.
How close are businesses to the self-driving WAN?
The rise of cloud applications and ever-increasing internet traffic has driven IT to evaluate augmenting or replacing their traditional network infrastructure, such as MPLS, with internet access. One approach is to break-out all internet destined traffic locally at the branch. However, in most cases enterprises require finer grained control and the understanding that not all internet traffic is equal.
A typical branch office will have flows destined to software-as-a-service (SaaS) applications that the business relies on, as well as flows to popular internet sites – such as employees doing home-from-work instead of work-from-home – and other unknown flows. Ideally, IT would like the ability to apply unique policies to each class of internet traffic; for example, making policies to send SaaS traffic via local traffic or MPLS, home-from-work traffic through a cloud-based firewall, and suspicious traffic through the full security stack.
Granular internet break-out policies sound ideal, but it’s difficult to accomplish, because a traffic steering decision must be made on the very first packet of the flow, which can’t be changed mid-stream. Traditional deep packet inspection (DPI) techniques won’t cut it, because the first packet of a typical connection often has no payload available for deep inspection.
To address this challenge and enable granular internet break-out, advanced SD-WAN solutions now offer first packet application identification and classification. First packet identification utilises a multi-layered learning architecture that encompasses learning locally in the individual edge devices – by snooping on domain name systems (DNS) and learning from DPI results – as well as learning at the enterprise level in the orchestrator by redistributing information learned by individual appliances (similar to fleet-learning for self-driving cars), and learning in aggregate.
The future of the self-driving WAN
Ultimately, enterprises are embarking on the journey to the self-driving WAN. Full autonomy is becoming a reality and, while SD-WANs aren’t yet fully automated, companies should watch this space as SD-WAN encompasses many advanced features. Indeed, by employing sophisticated machine learning techniques, such as first packet identification and classification, the self-driving WAN is getting closer.
Sourced by David Hughes, founder and CEO of Silver Peak
The UK’s largest conference for tech leadership, TechLeaders Summit, returns on 14 September with 40+ top execs signed up to speak about the challenges and opportunities surrounding the most disruptive innovations facing the enterprise today. Secure your place at this prestigious summit by registering here