Nebula gives us a glimpse of the future
Kevin Drinkall, EMEA Wireless and Cloud Market Development Manager at Zyxel, talks about the new features our cloud-based network management service, Nebula, which make it easier to connect new devices to the right virtual LAN, to get connected from anywhere securely, and make use of machine learning to ensure users get the best possible wireless experience.
We are always making improvements to Nebula, our cloud-based management platform. The latest additions give you a glimpse of what the future may hold for networking. We have made it easier to allocate devices and users to the right network segment and move them around. We are also using machine learning to give wireless users an even better experience.
The first new feature recognises a device manufacturer and allocates the newly-connected device to a specific virtual LAN, automatically. If you attach, for example, a particular IP camera to a Nebula-enabled switch, it can be identified and configured on the surveillance VLAN. Similarly, IP phones could be connected to the voice segment on the network.
This makes it very easy to connect a new device to a network – and perhaps even more importantly, to move them around and connect them to different Nebula switches. This is great from a security perspective, as it means it’s almost impossible for devices to end up on the wrong VLAN.
It’s also possible to do this with users. Different groups or even individual users can be identified and allocated to a VLAN. This means you don’t need to integrate with any external services, such as Active Directory, and there is no need to set-up multiple SSIDs within a wireless network; wherever a user connects to the wireless LAN, they will be assigned to the correct VLAN.
All this is set-up through the Layer 2 Isolation feature in Nebula and is done using the unique MAC address of every network-connected device. Once set-up, it and makes life a lot easier for everyone.
Better Wireless Health
In addition to these features, the Wireless Health feature in Nebula has been enhanced. Our engineers have used machine learning to make our access points even better at delivering the best possible user experience.
Our Nebula-enabled access points will constantly check the quality of signal between the antenna and each connected device. If a better signal can be found, it will switch the connection to a different band, antenna or access point. It does this seamlessly and without the user noticing any change. With the additional machine learning capabilities, it can now re-assign connections even more effectively. As it gains more data on what works well, it will switch to the best and most consistent connection sooner. The more it is used, the better it gets.
Just to be clear, this is about getting the quality and consistency of the connection right – not just the speed. Of course, it needs to be fast enough, but that is just one factor. If, for some reason, the connection keeps failing, being really fast does not help. It’s the user’s experience that we focus on, not just the speed.
Learning and adapting
Also, I’d point out that this is machine learning that we are using, not artificial intelligence. There is a difference: the former describes the ability of a machine to learn and adapt, based on data that’s collected over time. You could call it learning by experience.
Artificial intelligence is much more advanced and enables a device or machine to develop and apply knowledge, make decisions and adapt systems accordingly. It might, for example, not only add devices to VLANs, but also create new virtual networks.
Put simply, AI goes further; it is much more advanced and comes closer to simulating our own thought processes and powers of deduction.
The use of machine learning within Nebula is really just the first step towards us creating truly intelligent, auto-sensing, smart access points that will not only ensure every user and device always has the best available connection at all times, but will also auto-configure and manage devices, and keep the need for manual intervention down to an absolute minimum.
At some point in the future, there probably won’t be any necessity for anyone to touch the device, other than to remove it from the box, plug it in and turn it on. By automating the mundane and routine tasks that need to be performed to make networks function really well, we are saving everyone time and effort, and minimising the potential for errors or oversights. We are making networks that are built using Zyxel technology and managed with Nebula as efficient and cost-effective as possible for our customers.