Cloud-based network analysis drills down to users and apps

Nyansa's monitoring service can now identify applications and determine how the network affects them

Nyansa has added an app analysis tool to its cloud-based network performance service.
Credit: Andrew Hart

Cloud computing is throwing more variables into network performance, but it’s also helping to make network management smarter.

Nyansa, a startup that launched earlier this year with a cloud-based service to pinpoint the sources of enterprise network problems, is now adding the ability to analyze how individual applications are performing.

The company's original Voyance service uses a software crawler to capture data about how traffic is flowing over the network and how each piece of infrastructure – such as a router or a Wi-Fi access point -- affects that performance.

Now the crawler can identify applications, so it can relate infrastructure performance to each user’s experience with specific applications. It can also measure special performance parameters like Wi-Fi call quality.

Network administrators can drill down into those readings and find out if any network problems are hobbling an application. They’ll be able to examine how the app is affected by Wi-Fi, the wired LAN, the internet connection, and the device it’s running on.

20161104 nyansa voyance plus screenshot Nyansa

The Voyance+ service from Nyansa can identify applications and correlate each users' experience with network performance.

Armed with this new feature, Nyansa is expanding its product line with a service called Voyance+. It will cost about 50 percent more than Voyance, for enterprises, but current customers will get the add-on free.

There’s a growing focus in IT management on the end-user experience, and it’s coming at a good time, according to IDC analyst Nolan Greene. Both SaaS and wireless expose application performance to many more variables, so it doesn’t always work to gauge an enterprise’s IT health using conventional, internal measures.

Nyansa hopes to end frustrating conversations at enterprises about why applications aren't performing up to par. The way it is now, when an application is slow, APM (application performance management) tools may show that servers and software are running well, so network administrators are stuck trying to explain the slowdown. Often, they can’t.

"The applications have never been in the domain of the network guys," said G.T. Hill, Nyansa’s director of technical and product marketing. Voyance+ will give them eyes in that realm, helping them investigate and solve specific user complaints, he said.

The software can identify 218 applications so far, Hill said. For some, like Skype for Business and Cisco Unified Communications Manager, it uses APIs (application programming interfaces) for specific kinds of measurements. For some applications, including SaaS offerings, it uses third-party software to get application signatures. Nyansa can also identify custom enterprise applications as long as it has information like server names, Internet Protocol addresses, and TCP port numbers.

When it looks at applications, the network crawler keeps tabs on things like voice and video call quality. The data it collects is detailed down to the user, location, and device type. Nyansa also compiles metadata about application performance from multiple customers and analyzes it for broader insights, Hill said.

For example, Nyansa's machine-learning mechanisms have determined that the Pokemon Go app tends to perform worse than the average application in enterprises that have been testing Voyance+. They've also figured out why: Employees are using it in outdoor areas, where the companies' Wi-Fi networks are weaker.

Voyance+ can also tell how much time and data each application takes up on an individual user's device. If it's too much, administrators may throttle or disconnect the application.

The Voyance service starts at US$18,000 per year for a one-year enterprise subscription. Voyance+ starts at $27,000 per year. The annual price goes down with longer engagements.

To comment on this article and other CIO content, visit us on Facebook, LinkedIn or Twitter.
Download the CIO October 2016 Digital Magazine
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.