Government datacenter consolidation: deadlines more important than results

With a new study on Federal Datacenter consolidation, sponsored by NetApp, MeriTalk provides a compelling view of the difference between the private and public sector approach to datacenter consolidation. In the public sector, where every dollar spent has to be justified, consolidation plans include detailed analysis of the potential savings and explanations of how the consolidation will benefit business process.  In the public sector it appears that to the majority of datacenter operators, the approach is more of "it's been mandated, so it will happen, so why should I care about making it an efficient process or optimizing the results of the consolidation to build a more efficient environment?"

Perhaps this is simply another example of government inefficiencies; less than a quarter of the survey respondents (who were interviewed both in person and online) were able to track the savings found in the datacenter consolidation, a somewhat surprising number.  But even if you pass that off as specialized knowledge that only the personnel involved in those projects would have, you would also need to try to explain away the fact that less than a third of the government datacenter personnel surveyed could tell you the average load across their datacenter, which doesn't show well when compared to the 94% of private sector datacenter personnel, in similar roles, who were aware of their datacenter's workload.

The MeriTalk study highlights the fact that more than three quarters of the federal agency datacenter personnel surveyed said that they didn't know the PUE of their specific facility, despite the fact that, on average, 12% of the typical datacenter budget is spent on energy issues. The study also points out that this lackadaisical approach to datacenter management was an across the board issue; storage efficiency was also poorly monitored, with a quarter of the respondents not tracking storage efficiency at all.

After reading through the report it seems pretty clear that despite the existence of datacenter standards for management and measurement, there is no governmental oversight in the operation of these federal datacenters that requires their use. This makes it difficult to compare operations or to easily identify problem areas. It's also makes it appear that the old adages about spending other people's money are all true, at least in the federal datacenter community. While the datacenter consolidations appear on-track for the 2015 deadline, this study gives the impression that the entire process is pretty haphazard and inefficient.