Research: 2011 State of the Data Center
At the Crossroads: 2011 State of the Data Center Survey
Data centers represent a massive investment of fiscal and human capital. Given the choice between squeezing everything you can get from an existing facility vs. wiping the drawing board to start over, we understand the economic imperative to squeeze. And it’s clear from our InformationWeek Analytics 2011 State of the Data Center survey that most companies are opting to tweak the data centers they have rather than break new ground.
In this report, we’ll analyze survey results from 427 data center professionals to profile IT’s struggles to measure power and other resources, examine the tools and technologies being used (and not being used), and discuss ways that we can improve the efficiency of our current data center facilities. It’s clear from the results that there’s still a lot of work to be done in terms of measuring energy use and taking advantage of efficient products. For instance, most respondents have embraced server virtualization, but have yet to adopt IT products that can match power consumption to actual usage. And much more effort can be made around monitoring and managing power use.
More importantly, we’ll make the case for organizations to rethink how a data center should be built. Traditional designs, and the assumptions behind those designs, aren’t suitable for addressing the problems and constraints that our companies face today. New designs and new technologies are available, but mindsets have to be changed first. Companies such as Facebook are lifting the curtain on new data center designs. We’ll look at how its innovations in servers, electricity delivery, power supplies and cooling are upending traditional notions of how a data center must be built. We also make the case for greater information sharing among industry players, including IT organizations, equipment manufacturers, data center facility operators, architects and engineers. As more information becomes available to the community as a whole, we can generate widespread acceptance of groundbreaking designs.
New data center designs must be considered before organizations invest too much in fixing what we have. Incremental improvements to a broken system can be as wasteful as leaving inefficient designs and gear in place. If we don’t reconceive the data center, we will be stuck with yet another generation of “the way we always have done it,” with only marginally better efficiency to show for our efforts. And that doesn’t serve anyone. (R2120511)
Survey Name: InformationWeek Analytics 2011 State of the Data Center Survey
Survey Date: March 2011
Region: North America
Number of Respondents: 427