The Internet may be the most powerful business tool ever. But that power comes with a price. The Internet consists of, by some estimates, over 2,000,000 interconnected computers. And as is the case with any complex system, reliability is an issue. These hosts, and the hardware that connects them fail from time to time, and they all must be restarted periodically, resulting in temporary outages in parts of the Internet.

NIMS is our unique network watchdog system, tracking the status of all of our in-house network hosts, and the MCI computers that provide our Internet connection, instantly alerting the technical staff in the event of any major change in network condition. This helps our staff keep network downtime to as little as 0.2%.

How NIMS Works

NIMS works by constantly checking and rechecking the computers on it's watch list to make sure they are still active. It sends each host a network "ping" and waits for a response from that host. If the computer returns an "OK" response, then NIMS notes it as being active and moves on.

If, however, any single computer fails to give an "OK" response within a few seconds, NIMS swings into action, logging the incident, marking the host as "down" and immediately dialing the system administrators' pagers with specific codes that instantly tell them what the affected machines are, and what subsystem has become nonresponsive or entirely inactive. This and all pages sent by NIMS are logged in a file that allows us to later create detailed reports of system downtime activity.

NIMS then continues the check the rest of the hosts on its list. When it gets back to the down host, it will check it again. If it is still down, NIMS will ignore it for 10 minutes and then repage the system administrator. This keeps the administrator's pager from beeping constantly every 60 seconds while he works to resolve the problem, but lets him still be aware of the network condition.

Then, as soon as the host becomes responsive again, NIMS sends an "all clear" page to the system administrator, telling him that the host previously reported as down is once again responsive to network traffic.

Why?

We believe that NIMS, or some other equivalent network monitor, is an essential part of maintaining any reliable computer network serving commercial customers.

Would you buy a car that had no air bag OR seat belts? Would you force all of your employees to drive such a car? If not, then why would you stake your company's future on a network that doesn't have basic safeguards as simple as a network monitor? Remember that when your provider's network goes down, your customer will be calling you, not your provider.

In an industry where garage start-ups are common, the old saying holds true: You get what you pay for.

Other Benefits

NIMS gives us more than just a network monitor. NIMS makes its paging facilities available for use by other programs running on the network. This means that now all parts of the system have a means of sending information to a system administrator instantly.

The bottom line is that NIMS keeps our system administrators and our systems in touch better than anyone else.