The server on the far left (at the top of our rotary) exhibited almost 42,000 minutes of call time. It was more than 90 percent full during 80 percent of the day, or roughly 19 hours. The servers on the right are further down in the rotary and exhibit less usage both in call minutes and overall utilization.

In addition, you'll note that the second server has fewer call minutes, but the time shown at 90 percent is not proportionally lower. This indicates that both remote-access servers are heavily used throughout the day.

Server three logs 28,000 minutes of traffic, but the time spent at 90 percent capacity is less than one-third that of server one. This indicates that the network still has enough capacity to cover peak periods. If all of the graphs were peaked, Network Health would point out the need for additional ports.

The Service Level reports combine and present daily and long-term reports for various user levels, ranging from an IT manager to an executive. Each report provides both generalized and specific information about remote-access servers and groups. These accounts are useful for obtaining snapshots of workloads across servers and groups of servers.

In the lab, Network Health's Trend Analysis reports let me customize reports and dig for particular information. I scheduled a daily report covering the total number of errors for each remote-access server, as well as a weekly report for trend analysis. While Network Health reports may note problems, they are designed for preventative maintenance rather than for real-time troubleshooting.

For the complete Network Health experience, it's important to read the online documentation, which will help you to understand the assumptions behind the analysis. While Network Health-Remote Access offers simplified reporting, you will still need to watch the reported data carefully and adjust thresholds as necessary, as well as add to or modify existing reports.

Mike Fratto can be reached at mfratto@nwc.com.