Fluke released its newest version of OptiView XG v10, a network analysis tablet that allows simultaneous stress testing of the network along with diagnosis of problems. Fluke also added a new iPhone app, HeadsUp XG, that allows network administrators to receive notifications of network issues.
In my prior post: "Fluke Asks: What's Your Mean-Time-To-Know?" I addressed the significant amount of time that translates to big losses when companies' network administrators cannot isolate the root cause of issues within the network. We recently had a case in point:
We upgraded our FIOS service to 75/35 Mbps, but the process stalled because we couldn't break 35/9 Mbps. We removed the router from service and tested at the ONT (optical network terminal) to prove the bandwidth. Next we tested at the Ethernet interface between the ONT and our router. Then we reviewed the configuration and opened up a trouble call.
Over the next couple of days we removed RTP monitoring and made configuration changes as requested by the factory; each yielded little or no change to our throttled-back bandwidth. Subsequently, the support engineer remotely connected and then asked, "When did you last use Port Mirroring?" With the huge increase in bandwidth and with port mirroring enabled, the router wasn't able to process both of these efficiently. Once port mirroring was disabled, our bandwidth upgrade issue was resolved.
Dan Klimke of Fluke said OptiView would have pegged the processor occupancy of the switch, which would have helped us understand what was going on when the support engineer requested we disable RTP monitoring, since we observed a high CPU utilization (+90%) when testing bandwidth.
An added feature of the new Fluke release is the iPhone app HeadsUp, which provides instant access to network issues that I think is proactive, since network administrators can also drill down to the problems in the log.
This led me to discuss with Dan establishing baseline tests and then re-testing with a load on the network for new network turn-ups. This is pretty cool because with the network loaded and stress tested, network managers can establish a benchmark and then add a layer of change, re-test, and repeat the process.
This kind of benchmarking and testing should prove very useful. We've been busy in the education vertical, moving services over to the network. Adding fiber between closets and then swapping out old gear is just the beginning of many of these projects; adding voice with endpoints and then the SIP trunks comes later. Then there's layering on paging, bell schedules and ringing, and then WiFi--all of which bring challenges.
When the workstations are replaced, that's another substantial change, and having the ability to both test (load the network) and diagnose issues gives the network admin higher confidence.
Dan also mentioned the WLAN 802.11 a/b/g/n testing capabilities of the OptiView. OptiView goes beyond just testing; it comes with onboard spectrum analysis using the AirMagnet Spectrum XT option for RF analysis. This is important because non-WLAN devices can interfere with WLAN networks; and this got me thinking again about "the process."
The process almost always begins with a site survey and/or a plan. Using OptiView from the beginning and moving forward with additions, changes and upgrades makes sense, but it also eliminates a lot of guesswork and judgment calls that go into WLAN installations.
Admittedly this is a big sell in the SMB space once costs for the tools are outlined. On the other hand, I think it's fair to say there are hidden costs--costs for disruption and delay until problems are resolved, and then the actual costs associated with problem resolution. So then, what are the costs for not using a tool like OptiView, and what kinds of installation mistakes can be avoided, and how much more effective over time will network administrators become if they do use the tool?
