Careful, it's a trick question.

What if I also added that the first environment was an old, line manufacturing company with stagnant growth that is running on 8-year-old storage arrays and that the second environment is a young biotechnology company growing at triple-digit rates? Now the answer gets a good deal fuzzier. In fact, the answer is that you can't really tell which is better managed.

A point-in-time utilization statistic by itself doesn't convey enough information to paint a true picture. Storage capacity is dynamic, and in a smaller environment, a new purchase is going to have a negative impact on that point-in-time number. Even historic trending, while more useful, still may or may not provide an accurate portrayal of an environment. Additional subjective contextual data is required to render a judgment.

Improved utilization is an often stated goal, but I would suggest that it is actually a byproduct of a well-run storage environment, not the end goal or indicator. In the server world, virtualization has become a hot topic of late. Low server utilization, often on the order of 15 percent to 20 percent, is a factor often cited as a reason for moving to virtualization. In fact, the utilization rate means that there is bandwidth available to enable the consolidation efficiencies afforded by virtualization technologies. The true end goals are improved manageability and flexibility, reduced costs through consolidation, and improved levels of service through greater availability. Oh, and by the way, the utilization rate is now higher.

Determining appropriate measurements and applying them correctly is a big challenge in storage management. I have picked on utilization in this column as well a previous one, but there are others. Cost per gigabyte is a key metric that is just as likely to be miscalculated as it is to be misunderstood. Terabytes per storage administrator is a mostly meaningless statistic often employed to benchmark staffing levels.