A few months ago I drove solo through the Midwest. For 14 hours it was me, the open road, my iPod and nothing else.
Somewhere along Highway 70, a Bruce Springsteen song cued up from the shuffle, 57 Channels (And Nothin' On). It was originally released in 1992 as a social statement about the sheer volume of entertainment options without substance. It's not terribly old, but playing that day, it seemed so dated. I mean, come on, I get more than 57 channels with my basic cable package, and I could hardly take the song's message seriously with a chorus that was so out of touch.
A few miles down the road, I was on a conference call with someone from another state when the topic of data center consolidation came up. As I listened to my friend outline the need for a centralized data center under the CIO's leadership, all I heard was the chorus, "57 channels and nothin' on." The conversation suddenly seemed just as dated as the song.
A year earlier, while working as the deputy CIO of Missouri (a state with a consolidated data center for more than a decade), we conducted a study of the state's most critical applications. We did it for a disaster recovery and business continuity study, but what we learned was a lesson in the relevance of data center consolidation: 70 percent of our most critical applications are no longer on the mainframe or in the data center.
The Missouri Data Center is evolving. It has mature mainframe and print operations, a growing blade center and one of the best consolidated e-mail and archiving systems of all states. It also has customers who have moved systems out of the consolidated environment and into small server rooms scattered across the state.
The reasons why they built outside the center is a discussion for a different article, but the point is, even in a seasoned consolidation effort, we continued to struggle with the core issues of dealing with technology on an enterprise level, preparing for the future through the use of architecture standards, and exploiting the power of a common infrastructure.
Those are the issues typically at the heart of every consolidation effort. They are anything but dated, and we are constantly struggling with the best way to help our organization grow to be more efficient, secure and available to the people who have come to depend on technology as much as they depend on electricity.
The more I work with IT shops, the more I think Nicholas Carr was a genius when writing about IT as a utility in Does IT Matter? In the book, he compares factories and their waterwheels to today's organizations and their technology needs. The waterwheel provided electricity that operated the machinery inside factories. Similarly technology runs the factories of today's agencies. At some point, it made more sense to offer power to several factories as a centralized service from a single plant. IT is following suit, with the private sector moving at various paces and the public sector joining the mix. The concept is sound, but is this best done through data center consolidation? Are we putting our eggs in the wrong basket? Are we investing in the wrong type of power plant?
The idea of moving everything to one place certainly has appeal. Many argue that you can support much more hardware with much less staff, run a greener shop by reducing the HVAC requirements of multiple server rooms, and maximize the buying power of several agencies by coordinating their efforts. Others argue that a common data center means a single point of failure in the event of a disaster, workload priorities that don't always match customer demands, and multiple people who argue that they can provide the service more cheaply on their own. We spend so much time arguing over the "how" of data center consolidation that we stop talking about "why" it should be done.
We must change the argument over consolidated data centers to more fully encompass the meaning behind the effort. What we are after is a better way to utilize technology for our customers and, in turn, for the citizens they serve. Data consolidation might not be the argument for your organization, but an enterprise perspective, common infrastructure and reliable services are things that are as timeless as a classic song, like Springsteen's Born in the U.S.A.
NEW ON THE PODCAST