Self-driven networks are networks that combine telemetry, workflow-automation, DevOps and AI. By combining all these techniques, the network can adapt in a responsive way to the changing circumstances.
Besides this, a self-driven network can also detect proactively certain problems and changes in the network to avoid problems and interruptions.
The concept self-driven network is not only about the network. The user experience is key. That’s why the infrastructure needed to manage the network can’t be an issue anymore. This is not the part where most of the resources of the IT-team should go to. Because of this philosophy a self-driven network is fully Cloud-based in terms of management and monitoring.
A self-driven network also isn’t dependent on the infrastructure needed to manage it. It doesn’t matter if the network is small of large, the infrastructure scales when the network is growing.
A self-driven network is built on multiple aspects:
As a result you actually get a continuous cycle. Data enters the cloud (network traffic, logs,…) and the network analyses this data. Based on this analysis, the network will make certain decisions (optimisation of the network). This way you get a new network configuration (e.g. in case of WiFi adjusting channels, power,…) which will generate new data to be analysed by the network.
Juniper Mist has a virtual assistant that collects the data from the network, processes it and creates a number of actions/alerts based on it. It is also possible to ask the virtual assistant questions such as:
There is also a very powerful query language that makes it easy to perform troubleshooting sessions. It also offers packet dumps that allow you to see in detail which packets were passing through the network at the time of the problem.
Juniper is also working on integrating the rest of its portfolio (EX switches, Session Smart Routers and Firewalls) into the ecosystem in addition to wireless and, in this way, including data from all network layers in the algorithm.
Extreme Networks is already known for their CloudIQ ecosystem where access points, switches and 3rd party devices (monitoring via the XIQ Site Engine) can be managed. Recently it also made the CoPilot available.
The CoPilot processes all data using Machine Learning and AI and then shows in one page what possible problems are on the network. In this way, it is faster to find what possible problems are and easier to troubleshoot.
Extreme Networks is also working on the AutoPilot which also solves problems by itself when needed. This way it is a self-healing product.