Fluke claims fix for Cisco WAAS issues

The latest version of Fluke Networks' Visual Performance Manager can diagnose and fix performance problems in Cisco's WAAS application acceleration, as well as measuring the ROI achieved by WAAS, Fluke has claimed.

The software also applies expert guidance and rules to help network administrators get to the root of a problem faster, and provides red/yellow/green dashboards or "heat charts" to show executive managers how problems affect the performance of critical network applications, the company said

VPM version 5 can compare the accelerated and non-accelerated performance of an application, and drill through to see exactly how WAAS is accelerating WAN traffic, said Fluke business development manager Doug Roberts. It has some similarity with technology from NetQoS, he noted.

"WAAS offers four different levels of optimisation - TCP flows, data redundancy elimination, compression and specific application optimisations. So depending who you're talking to within Cisco, they will tell you WAAS is not the easiest thing to set up," he said.

"There are some guidelines from Cisco on the levels of improvement to expect for different application types, for example CIFS, HTTP or MAPI. We can verify for the customer that they are receiving the expected percentage improvement or better.

"And if they're not receiving those, we can dig down to the transaction level to understand the actual policies being applied on a per-user, per-transaction basis."

Fluke has been working with Cisco on this technology for around six months, and recently started a similar project with Riverbed, said Roberts. "We will most likely start work with Blue Coat next - they've been brought to the forefront by the Packeteer acquisition," he noted.

He added that VPM 5 helps admins troubleshoot problems faster because it applies expert guidance based on working with existing users.

"We worked with customers," he said. "We asked 'What do you think would be the next most logical approach if this happened?' So it uses deductive reasoning - you click on a spike, say, and it takes you to the next level of information associated with that spike."

He claimed this makes it possible to identify the root cause of an application slowdown in three clicks or fewer.