Few people think about the types of alerts they receive. Yet, having the right types of alerts configured can result in significant cost-savings, improved infrastructure performance and increased end-user productivity. By planning and categorizing the right alerts for your company’s infrastructure operations, you can be confident that your IT operations are running smoothly and efficiently.
Alerts can be categorized into three classifications:
Performance Alerts – These are simple alerts used to make sure day to day operations are running smoothly. These are things that may need immediate attention, such as a sudden spike in CPU utilization that’s causing your end-users to have a bad experience, or a large memory increase indicating that something needs to be investigated in the very near future.
Capacity Alerts – Classic alerts that warn end-users about when storage is full, CPUs are at full utilization, RAM is at zero or other factors. Capacity alerts have evolved to include triggers before capacity thresholds are exceeded. These triggers alerts IT teams when things should be fixed or fined tuned in order to reduce operating costs or increase productivity.
Dynamic Alerts – A modern, smarter approach to notification is the dynamic alert. A dynamic alert has the ability to move the threshold parameters based on day-to-day or week-to-week use. Say on a Monday morning, CPU utilization of your farm is high at 80% due to workers showing up after a weekend, ready to tackle their assignments. A dynamic alert would be smart enough to capture typical Monday CPU utilization and classify that use as “normal”. Yet, a dynamic alert might be triggered if CPU use on a Monday hits 99% or drops to 30%, a clear sign that there’s a problem. By analyzing trends over time by time of day we can create intelligent alerts and become aware of things which otherwise may not be noticed.
Now that you know the three types of alerts, you should look at your current alerts and see where switching alerts to a different type might help decrease the amount of alerts you get while improving the fidelity of alerts that need immediate attention. Good luck.