Digital Twin

I was very fortunate to be a delegate at Networking Field Day 13, all the way back in 2016. This was a milestone event for Forward, as this was their official move out of “stealth” as a Silicon Valley startup. Their initial presentation was impressive, and the Forward Networks platform offered something I had not seen before, an accurate digital copy of your network, which you could query to understand paths and flows, and test proposed changes to prevent misconfiguration.

Fast forward to Networking Field Day 34, and Forward Networks are presenting again, now to highlight the maturity of their product, and how they are now integrating AI and LLM to allow for an even better experience with natural language queries, and enhanced visibility into your network digital twin.

At it’s core, the Forward Networks platform remains a full digital twin of your environment, allowing you to search, query, verify, and predict how traffic is behaving, and will behave in your environment. They are vendor-agnostic, meaning you can easily have a mix of Cisco, Juniper, HPE-Aruba, Arista, etc. and still leverage the power of the platform. A simple local agent crawls your network with SSH credentials (or via API if you have devices that don’t support SSH) and builds the snapshot of your network, which you can then import into the tool, and begin working with.

Having evolved quite a lot since 2016, Forward Networks now included integrations with the 3 major cloud providers, and security tie-ins to platforms like Rapid7 and Tenable to identify CVEs that may impact your network devices. Now they have taken the next step, with integration of Generative AI with AI Assist as part of the Network Query Engine (NQE).

AI Assist now allows the use of AI LLM to generate queries against the network model. These queries can be saved for use later in your own repository of queries, or you can also use a number of pre-packaged queries out of the box. The reverse is also true, and you can use Summary Assist to analyze a query and provide a plain language summary of what it is doing.

Proving the Negatives

If you’ve been in networking for any length of time, you know the feeling of having to “defend” the network because it’s the first thing that gets blamed when something isn’t working. We’re constantly having to prove a negative, which is sometimes hard to do. It can involve a lot of jumping around your network in the CLI, pinging and checking routes, doing packet captures, etc. and there’s no easy way to translate a lot of these methods into a simple to understand view of your network, and where the traffic is or is not going.

The Forward Networks platform provides a simple, easy to understand analysis and view of traffic flow across your network in a 100% mathematically accurate carbon copy. Queries can be copied and shared, so now you can send a link to your Dev team and show them that, despite their initial assessment with no troubleshooting or factual information, it is *not* the network.

Contining Forward

The team at Forward Networks continue to evolve and strengthen their platform, and the integration of AI LLM with AI Assist and Network Query Engine is a perfect fit. In an era where everyone is trying to shoehorn AI into their product, whether or not it makes sense to, this is an excellent example of what is still a very immature technology, put to good use.

If you want to learn more, and check out a customer testimonial around automation and cost savings from one of Forward Networks’ biggest customers, you can watch the recordings from the presentations here.