Preface

Have you ever had complaints from your customers about poor network performance? What about trying to find out what your bandwidth utilization is from the edge? If you are an IT administrator, I guarantee that you have had these types of tasks before.

I recall a time when I was an IT administrator of a medium-sized business, working at the company headquarters. The business had a data center hosted in Little Rock, Arkansas with more than twenty different branch offices scattered throughout the United States. The data center was the central hub for all network connectivity for the entire organization with each branch office connected to the data center via private MPLS circuits. One day, my team received a call notifying us that one of the remote locations was without Internet access. There was literally no way for me to know why this happened without spending a great deal of time researching the issue. After an hour, we finally found the cause of the problem. The core router died during a lightning storm at that branch office. We were able to call a local technician to connect a spare router at the branch office and get our customers back online but the damage had been done. The total amount of downtime for our customers was four hours which was completely unacceptable for a company that relies on the Internet to perform its work.

In a completely different example, I was working as a network administrator team member at another company where most of our users use a web-based application to perform their jobs. One day, I received an e-mail alert notifying me that our primary Internet link was down. I contacted our ISP who dispatched a technician immediately. As I was working on crafting a notification e-mail message to the company about the situation, I received a call that many of our customers' Internet connections were very slow, their web application was timing out, and they were unable to work. I informed the customer that we were working on the issue and notified the company of the problem. After a short period of time, the ISP technician arrived and resolved the problem. The total customer downtime for this scenario was 30 minutes.

As you no doubt have observed, there are multiple issues with the first scenario. There was no alerting in the event of any type of network failure which limited the IT department to be proactive in such an event. The second scenario shows some of the best and most used features of a network monitoring system. Thanks to the core monitoring features of the monitoring system, I was able to determine the root cause of the problem quickly and have the ISP technician dispatched as soon as possible. Even though my customers experienced a network outage for 30 minutes, I'm sure you would agree that a downtime of 30 minutes is more acceptable than four hours.

SolarWinds Orion Network Performance Monitor is one of these types of monitoring systems and this book is going to discuss many of its features including what Orion NPM actually is, what it does, the technologies behind Orion NPM, and how Orion NPM can help to make your job as an administrator easier.