Network management can be an esoteric topic that rarely yields a sexy headline. Given the difficulty in managing application performance over temperamental unlicensed airwaves in a wireless LAN, however, the subject can quickly become paramount once the complaints start rolling in. In this spirit, Meru Networks claims the ability to slash Wi-Fi troubleshooting time with a new correlation engine and, optionally, the Apple iPhone.

The company this week announced version 2.0 of its E(z)RF Network Manager, which Meru senior vice president Kamal Anand says prevents users from slogging through event logs to handle trouble tickets. It does this by continually capturing each over-the-air event for every client in the network, storing the events in a central database and correlating them in what it calls an Inference Engine.

This allows the system to quickly and accurately deduce what's up compared to the traditional way of viewing only summarized data, Anand says. The company cites case studies showing an 87% to 91% reduction in the time it takes to resolve a wireless problem.

"The same wireless symptom can have many causes," Anand explains. For example, if a network manager sees a client device making a number of associations with an AP but receiving no IP address assignments, the engine might infer that the Dynamic Host Control Protocol server is down, he says. With only periodic, aggregated statistics, it might take a human much longer to get around to this diagnosis.

Users can access the system from a traditional console, an Apple iPhone or an Apple iPod Touch. The client system is browser based and isn't sold at the Apple iPhone Apps store. But Anand says the system dashboards were specifically formatted for the Apple platform and that it "looks native."

Meru says the company picked the iPhone as its first smartphone platform because many of the company's customers are higher-education organizations, in which the iPhone is an entrenched tool.

Because the system can show a snapshot of a client's state at any point from the present through the past, this enables IT staff to also troubleshoot past events and take corrective action, Kamal adds. For example, the system can determine that the network becomes overloaded at a certain time of day in a certain location, and IT can take steps to increase capacity in that location.

Joanie Wexler is an independent networking technology writer/editor in Silicon Valley.