SNIA data storage innovation conference: the lies of your performance metrics

By Barry Cooks, VP of Engineering –

This past week I had the pleasure of speaking at the SNIA Data Storage Innovation Conference about one of my favorite industry misconceptions: the idea that all supposed IT performance metrics are created equal. Given that I work at an infrastructure performance management provider, I have a strong respect for the value that performance metrics can offer, but that means I also have a first-hand understanding of which ones are really worth an IT team’s time versus which ones are better left out of the picture. Unfortunately, we far too often see enterprises getting caught up in the second category and finding themselves unable to access the more beneficial statistics.

Where common performance metrics go wrong

This can happen for a variety of reasons, the most compelling of which is simply a lack of resources. The less impactful metrics, such as error counters and IT utilization, have traditionally been the easiest to gather, making them more appealing for IT teams that are often already overworked and understaffed thanks to the current economic climate. They are regularly being told to do more with less, so the idea of focusing on quick and easily obtained statistics is appealing as it allows them to focus on more time intensive initiatives. Unfortunately, error counter and utilization metrics don’t really tell you anything about the performance of an infrastructure, and the resulting averages don’t paint very clear pictures of the true highs and lows during that time period.

How you should really measure your IT infrastructure

Now, you’ll notice I said these metrics have traditionally been easiest to track down. You now have the opportunity to break with tradition. Legitimately valuable performance metrics are no longer the pipe dream they once were. Working with anything more granular than averages was previously unattainable just due to the imposition that kind of initiative placed on infrastructures. No one wanted to risk a negative impact on their production workloads, so the least invasive metrics were the least risky; however, as I covered in my presentation, modern infrastructure performance management (IPM) platforms eliminate that risk and offer the in-depth visibility IT teams truly need. Rather than settling for the easiest metrics, it’s time for global enterprises to start leveraging the most valuable metrics. Then, and only then, will IT troubleshooting become a thing of the past and optimal performance across the board can become a guarantee.

Interested in joining our team at any other upcoming events? Head to our newsroom and scroll down for our calendar.