Published: 12 Jul 2004
Sexy IT chatter gets executive attention from both the people who develop technology and those who use it. Right now, information lifecycle management (ILM) is hot, hot, hot. Grid and utility computing are also huge draws. People line up to hear about cool stuff that will neither generate anyone any money nor help IT in any near-term way.
Meanwhile, the reality that nobody wants to talk about at the executive level are boring old things such as backup and restore, which are still causing endless real-world problems in 95% of IT shops. The problem is worse than it was five years ago. And It will probably be even worse next year.
Show me another area in your life where it's OK that something you do fails 40% of the time. Unless you are baseball manager, if your employees fail in their tasks that often, you fire them! Yet somehow, we not only tolerate that level of backup/restore failure, we are afraid to upset the broken apple cart. We are held hostage by our people and their dysfunctional processes because we're afraid that if we make them mad and they leave, it might get worse. How dumb is that?
Outside of EMC and IBM, the systems vendors don't get it either. StorageTek, who arguably should own the backup/restore space, thinks backup is passÉ. HP and Sun, who have a massive installed base to protect, don't even pay it lip service, thus letting Atempo, BakBone, CommVault and Veritas grab business. Sun still sells Legato--I'm not sure it realizes that EMC now owns Legato.
Vendors don't seem to realize that despite all the attention on ILM, there's still no budget for it. There is a huge budget for data protection--including backup/recovery and disaster recovery. They continue to be the top operational problems inside IT.
Imagine if vendors invested some effort in creating technologies and services that actually improved the recoverability of data. I'm no rocket scientist, but I'm pretty sure I could sell that: "Mr. Customer, I can show you how to take MegaHuge Corporation's data recoverability levels from 40% failure to 20% failure for a million bucks." I think the customer would cry because they would be so happy. Of course, that means they still have completely unrecoverable data 20% of the time, but let's not wreck the beauty of the moment.
Backup/recovery is about to become sexy again. Traditional backup/recovery methodologies simply don't work, and won't work in the future. We need a new way to skin that cat. Will the continuous capture guys win the backup war? Probably not, but they have a fantastic disaster recovery methodology. Will virtual tape systems crush the existing tape industry? I doubt it, but it certainly should have a place at the table. Now is the time to look at some very cool emerging technologies coming out of startup land. These problems can be solved.
In summary, we have a perfect storm of stupidity happening here. IT executives don't pay enough attention to fixing the backup/restore issues that we all know they have. They accept absurd failure rates. They put businesses at risk. They need to take their head out of the sand and spend some money and make the risk smaller, not let it continue to grow. The vendors aren't doing anyone any favors by ignoring the space in lieu of the hot topic of the month. They are missing the opportunity to solve some real problems and make some real money. Boring, but true.
- Demystifying storage performance metrics –ComputerWeekly.com
- Containers and storage 101: The fundamentals of container storage –ComputerWeekly.com
- A guide to hybrid flash storage arrays –ComputerWeekly.com
- All-Flash: The Essential Guide –ComputerWeekly.com