In late 2014, Avi Networks came out of stealth mode with a product aimed at disrupting the application delivery controlled (ADC) market. Network World's Jon Gold did an excellent job covering the launch and the way Avi is attempting to differentiate itself, so I won't rehash what he has already covered.
In the right environment, the value proposition of what Avi is doing should be obvious to anyone covering the software defined networking (SDN) or network functions virtualization (NFV) market. Avi brings a high level of agility to the ADC, enabling customers to deploy ADC resources anywhere they need to in the exact quantity required. The pay-as-you-grow model means organizations are no longer required to overpay for resources they won't need 90% of the time. Instead, they can provision for normal utilization and then purchase more capacity when the workloads require it.
There's another aspect to what Avi is doing that can be very beneficial to the network operations team. Avi's analytic capability can shorten the "mean time to prove innocence" for the network by being able to rapidly zero-in on what resource is actually causing the problem. In the court system, you're innocent until proven guilty. In IT, though, the network is often guilty until it can prove its innocence.
As a former network engineer, I know all too well what rolls down hill and, unfortunately, when you look at the IT stack, the network is always at the bottom. Database responding poorly? Must be the network! A VoIP call can't go through? It's that crappy network again! How about the poor-performing CRM system? Clearly it's the network's problem. When I was a network professional I felt like I spent more time proving that a problem had nothing to do with the network than I did actually troubleshooting real networking problems. I refer to this as "resolution ping pong," as trouble tickets would get bounced around from the application developers to sever ops to a database administrator until it got to the network.
Avi has a unique solution that can put an end to resolution ping pong. In addition to providing ADC capabilities, the company collects a wealth of in-line data and has analytic capabilities that it displays in a dashboard to give a complete view of what's going on in the environment. ADCs sit at the junction of applications, servers, and the network, so they are in a unique position to understand exactly where an application problem originated.
Avi uses this position to collect application, user, server and network data. The information is de-duped, filtered, compressed and then stored in a real-time, streaming data store run through an analytic engine, with the data displayed in an intuitive dashboard. The dashboard displays end-to-end timing, application health scores, baselines, trending information, and a wealth of other data that can be used to monitor and troubleshoot application problems.
Avi enables administrators to query the data through Google search functions and can troubleshoot issues down to a specific processor or even software versions on different devices. For the networking team, Avi provides the necessary data when dealing with the various IT teams to prove that the problem isn't in the network. As they say on infomercials – but wait, there's more! Instead of taking the trouble ticket and ping-ponging it back to the application team, Avi can be used to isolate the problem.
ZK Research data indicates that 90% of the time taken to fix a problem is in the identification phase (disclosure: I am an employee of ZK Research). The long lead time in problem isolation stemmed from the fact that there was no quick way of looking at IT from end to end. However, big data analytics, SDNs, and tools like these are finally changing the game.