IT Performance and Availability monitoring can be a complicated and expensive endeavor. Getting to the point where it is all worth it can be a challenge. Unfortunately, many companies (sometimes even very large companies) tend to get as far as basic availability monitoring and then stop there because the additional steps appear to be too complicated. This essentially turns a potentially multi-million dollar investment into little more than an automated tattletale. Even worse, many times proper monitoring hasn’t been implemented that can even legitimize the rest of the data being collected so the result isn’t just a tattletale, but a tattletale that is giving inaccurate results. Here’s 3 ways to get over those hurdles and get some real value out of your investment:
1. Use performance tools as performance tools
This may seem a little too straight-forward, but the fact is that many end user experience performance monitoring tools are billed as availability tools right out of the box. Measuring availability is trickier than just tracking a failed transaction. The problem is that these tools are typically sending out a limited amount of probes so the resolution of the outage is coarse. A single bad script result can show as a 10 to 15 minute outage. That is rarely the case when there is only a single bad script result. There are many transient issues in the network that can create a very temporary issue. While it is a problem that should be addressed, each problem shouldn’t count as a full-blown outage.
2. Don’t cover up real issues to compensate for impossible expectations
The problem is that, many times, companies will try and force a tool to work in a way that is contrary to the very core of its design. If a tool is intended to capture performance and problems, but not availability, then don’t cover up the ability to capture problems just to get availability. You’re losing out on the most valuable features of the product that can help you to identify trends to prevent issues before they impact end users. For example, if you’ve covered up the single transaction failures just to make a tool give more “accurate” availability data then you won’t be able to see when those transient errors become more frequent and eventually turn into a real outage.
3. Don’t put all your eggs in one monitoring basket
If a tool is made for end user experience monitoring, then use it for that. However, you’ll need more than just that side of the picture to do proper monitoring. You’ll need to invest in proper server, hardware, and network monitoring in order to get the full picture. These investments don’t need to be significant. The supplemental monitoring types can be filled by many inexpensive or even free monitoring tools. Also, when looking for these monitoring tools, ensure that they can provide a fair amount of historical data for trending analysis and historical reference when performing root cause analysis tasks on production issues.
About the Author
Matthew Bradford has been in the I.T. Performance Business for 13 years and has been critical to the success of many Fortune 500 Performance Management groups. He is currently the CTO of InsightETE, an I.T. Performance Management company specializing in passive monitoring and big data analytics with a focus on real business metrics.[contact-form-7 404 "Not Found"]