We all know the old saying, “The more things change, the more they stay the same,” but it is interesting to see just how well it applies in enterprise IT. For example, the recent set of third-quarter server sales figures from Gartner demonstrate the verity of the old saying rather well.

If the exponents of change are to be believed then Risc processors are old hat and are just working their passage towards obsolescence. The same has been said about Unix. Yet Gartner’s figures show that a big slice of the IT-buying community has obviously not been listening. IBM’s sales of Unix servers – which basically means its pSeries boxes running the Power family of Risc processors – jumped 12 percent in the third quarter.

What is more, though IBM shifted only 60 percent of the number of servers sold by volume leader HP, IBM’s server revenue was over 30 percent greater. So it seems that, from the buyers’ point of view, big iron still wins. To amplify the point, Sun Microsystems’ Unix server sales – seen by many as a benchmark of that company’s impending doom – rose 20 percent.

So what is going on? Well, I suspect part of the answer can be spotted in some of the other figures collected by Gartner, such as the poor performance of those server vendors offering Itanium-based boxes. HP’s sales of high-end servers slumped 21 percent, while Fujitsu’s tumbled by 10 percent and Bull’s fell 12 percent.

One interpretation of these trends is that the user community sees change in vital back-end infrastructure as a bad thing. But it is also not unreasonable to suggest that the early iterations of Itanium brought nothing to the party to tempt either new or existing users away from more established platforms.

But there could be another reason for the continued adherence to old iron – albeit the newer iterations of old favourites. This is the notion that the hardware platform is no longer relevant. Well, of course it is relevant in that it has to exist, but its type and nature is no longer so pivotal. Significant changes to hardware platforms, with all the disruption and service dislocations that can ensue, are no longer needed.

This does not mean that change no longer happens, of course, but it does mean that the focus of change has moved on. The point of consequence today is the overall enterprise infrastructure and its management as an interconnected entity.

So long as a server can be managed within and as part of the overall infrastructure, it can have a role regardless of the newness of its technology. Indeed, factors such as reliability and the experience of support teams in keeping systems running now point to “old” technology being the more profitable option for the enterprise as a whole.