At the very least, the views to data need to be centralized, but companies found this year that managing for Section 404 of SOX, or getting back on track after hurricane damage, or just setting up an archiving system is easier if the data is all in one place. I'm not saying we'll see a wholesale shift, a re-centralization, if you will, of all those distributed servers out there.
But IT organizations have started to learn that the data that counts needs to be counted, and putting it in a secure place where it can be viewed in context of all the reasons we now need to prove the existence of data is simply the best way to achieve that.Companies are also finding that optimization is more efficient in a centralized environment and they can do more with less.
So, while the systems and applications are still largely distributed, the data is getting centralized, normalized for reports, backed up, archived and viewed in a common location. With this method, the cost of data audits and access management also fall from their 2004 and 2005 levels. Don't mothball those mainfames yet.
As I mentioned at the top, this is prediction number four in my 2006 list. If you missed the first three, check back on: