By: TomLiotta

TomLiotta — Thu, 06 Sep 2012 11:25:33 +0000

How many data centers are involved? Considering the likely high number, I’d say the question is almost meaningless. The answer is “Some are; some aren’t.” The reasons going both ways probably number almost as much as the number of data centers.

I have experience that might help illuminate an aspect of difficulties faced by government agency data centers (and government agencies in general.)

I worked at the municipal, county and state levels for almost 15 years. During the final six years, I worked for a state-level Department of Education. The timing had me starting just over a month before election of a new Governor (and various Legislators, too). I then stayed through that Governor’s term for four years and then two years into the next Governor’s term. (And changes in the Legislature.)

In six years, I went through three administration changes. In the months after each change, the general atmosphere pushed by top officials was shaped by “We have a ‘Mandate from the People’. Things are going to be done differently now that we’re here.”

The common result was chaos. Many things had to stop as agency units worked on plans to implement whatever ‘new, improved’ ideas were mandated. Of course, in many cases, the basic pattern was to dust off plans from three, four or more administrations from the past and freshen them up with minor twists. It didn’t matter that they didn’t work well in the past, nor that they were just useless today. They were effectively the same old ideas, just brought back to life by new bosses.

So, whatever structures were previously implemented or were possibly actually being bent into some kind of working condition, it was time to start tearing them out and replacing them. Getting to know some of the people who’d been in service for a couple decades and more taught me how tiring and frustrating it was.

To the point of the question, what’s the definition of “efficiency”?

Many would think in terms of energy costs and capital costs. It’s harder to think of the human costs of implementation and harder to quantify them. Yet, the human element is a fundamental requirement for any lasting success.

The top is always subject to the whims of the voters who commonly have no idea how things work. With political offices at the top, the new directions from the top are often contradictory to directions that are currently being followed. Past costs may be sunk costs, but efficiency has to be thought of as going down when costs are regularly written off. And those who are responsible for, or simply work in, the affected data centers will be part of the costs. Some centers will close; others will absorb their work. Either way, it will be expensive in human terms.

Nevertheless, it’s all based in our system of representational democracy. I, for one, am not willing to give that up. I can accept that a definite degree of inefficiency is a natural consequence of what we have.

Some agency data centers operate efficiently. Others don’t. Change is to be expected. Data center consolidation can increase operational efficiency to a large degree.

At the moment, it doesn’t matter if this is a good or bad thing. The concept has sufficient approval and support to see it through. With competent staff committed to the projects, it could even have significant long term benefits.

Just don’t expect it not to be expensive.

Tom

Comments on: Are current data centers able to run efficient enough?

By: TomLiotta