X
Tech

100 datacenters here, 100 datacenters there, 1000 datacenters where?

With a short term project requiring a major federal datacenter consolidation effort, the discovery that there were twice as many datacenters involved gives a whole new meaning to "close enough for government work."
Written by David Chernicoff, Contributor

In a story by Rutrell Yasin of Government Computer News it was reported that the US government has discovered there are almost twice as many datacenters as previously thought that will need to comply with the Federal Data Center Consolidation Initiative. Based on data provided by the OMB, there are 2094 datacenters to be consolidated, a significantly higher number than the 1100 initially estimated by that same agency.

I have to admit that when I read that I thought that the definition that they were using for datacenter must be rather generous, with some agencies reporting a couple of unmanaged servers sitting in an office closet as datacenter assets that need to be consolidated. But further reading found that the definition for a datacenter for the OMB means a single room of at least 500 sq. ft., dedicated to data processing, which meets one of the Uptime Institutes four Tier levels for datacenters. So we aren't talking departmental servers hiding in a closet here, but rather actual data processing facilities.

This really does seem to indicate some massive inefficiency in the data processing model used by these government agencies. And the potential for not simply the energy efficiencies demanded by the government initiative but the opportunity to revamp and streamline any number of governmental processes. But given that the OMB is scheduled to approve agency consolidation plans by December 31st it seems unlikely that given the overall short timeframe and the complexity of dealing with government culture (including the military) that the submitted plans (filed by August 30th) are little more than the simplest physical consolidation plans that could meet the guidelines laid down for the agencies.

Remembering the IRS computing upgrade fiasco of the late 20th century I would have hoped that the government had learned from its previous mistakes. But given the short timeframe and unrealistic goals of this consolidation effort, it would seem that no one was paying attention.

Editorial standards