In a middleware system like BizTalk, it’s important to have a quality monitoring program in place. Monitoring of the BizTalk environment makes all the difference between an organization that runs efficiently and proactively prevents issues and one that seems to be constantly putting out fires. How many times in your organization has an application update caused a crash and everyone immediately blames BizTalk as the culprit? Proper monitoring reduces incidents in your BizTalk environment, eliminating the finger-pointing before it starts.
So, what are some of the signs of poor monitoring?
Frequent downtime is a big red flag. If you’re constantly getting calls at 3 a.m. because customers are unable to access your company’s website, poor monitoring is the culprit. Why? It’s because automated monitoring identifies anomalous behavior in your environment, alerting you beforehand of activities that could potentially become big events.
Budget-busting hardware costs are also an indicator of poor monitoring. If your IT team is frequently restarting servers and host systems, and you’re continually wearing out equipment, automated monitoring is the answer. By monitoring serves and business processes, an automated monitoring solution is able to identify when applications are disrupting operations. Equipped with this information, your IT is able to better optimize resources and streamline operations for better efficiency.
Manual techniques — no matter how good your IT team is — are inefficient as a monitoring solution. Amid the thousands of processes occurring in your organization, it’s impossible to determine the best parameters for every activity. Manually setting thresholds based on these parameters is an even bigger task, and ultimately prone to inaccuracies. The problem with manual monitoring is that you end up with a lot of noise. For example, depending on your configuration, a suspended message may generate two or more alerts for the same problem. And even then, it’s still too easy to overlook an important alert buried in all the noise.
The truth is that your organization’s IT team is made up of professionals whose talents are being wasted if all they’re doing is digging through alerts. It’s also likely you have a long list of projects you’d rather the IT team was working on. If you have too many projects on hold and a chronic shadow IT problem as a result, your monitoring efforts need to change — and fast.
One of the biggest signs that your monitoring is insufficient is when end users identify problems before IT is aware of them. No IT professional wants to walk into a firestorm started by customers calling in because the website shopping cart isn’t working. Not only does the ensuing scramble add undue pressure to an already overworked IT team, the situation also means you have some explaining to do. If you’re getting more than your share of loud lectures, you need to look at a better monitoring solution.
Although System Center Operations Manager (SCOM) is a powerful tool for monitoring BizTalk, it’s a limited solution. Microsoft is committed to evolving BizTalk, but that evolution doesn’t seem to extend to SCOM. Also, SCOM is not self-learning; it doesn’t automatically evolve as changes are made in your environment. Instead, SCOM requires your team to define parameters and set thresholds to generate alerts, which a tedious, time-consuming and inexact process. Why are you using manpower to do this when there’s an automated solution that is better able to handle the task?
The whole point of monitoring is to be proactive and move from a perpetual firefighting mode to a fire prevention mode. If your team is so busy reacting to alerts that it doesn’t have time to monitor system performance, then you need to look for a better monitoring solution.
Your organization made the decision to integrate and automate tasks by deploying BizTalk, so why is it using time-consuming and costly manual techniques to monitor the environment? Don’t your IT professionals have better things to do with their time?