"It's All Taken Care Of" -- Or Is It?

Posted by B. Boyers on Thursday, June 12, 2008 - 01:49 pm:

Share | Create New | Reply | Edit

We’ve all had times when someone told us “it’s all taken care of” and it turned out not to quite be the case. For example, you hire a plumber who ends up spending the better part of a day under your house and reports “it’s all fixed,” but several days later, at about one o’clock in the morning, you discover that your brand-new living room carpet is now soaking wet and ruined from a geyser coming up through your floor. Or, in the middle of a heat wave you spend a small fortune having your air conditioner fixed; the next afternoon you wonder why you can’t stop sweating and discover nothing but hot air blowing from the vents.

This seems to be just as common—if a bit more subtle—in the business computing world. An IT personnel reports to the IT director that the new OS patch was deployed—yet help desk calls continue to come in with the very same complaints the patch was supposed to eliminate; turns out the IT person was on the phone with a buddy while supposedly deploying the patch and it never quite made it out.

It can become even more insidious when a solution has actually been put in place and everyone assumes that solution is taking care of the problem. A prime example is scheduled defragmentation. On the surface it make sense: a defragmenter can be scheduled to run during off-hours so the file fragmentation doesn’t affect performance. But then complaints of slow performance continue to come in, hard drive reliability problems continue to occur and backups still seem to be taking far too long.

A closer look will reveal several major problems with this solution. First, scheduling time on a busy server for defragmentation has become a challenge, especially with a server that must always be up and running—a very common case these days. Second, because of today’s enormous file sizes and huge disk capacities, fragmentation is worse than ever and it continues to compound and impact performance in between scheduled runs. And lastly, there are cases (as with very large drives) that fragmentation isn’t being addressed at all.

If you want to put in a defragmentation solution that can be truly reported as having “taken care of it,” you have to turn to one that is completely automatic, that defragments invisibly and consistently in the background. Because only otherwise-idle resources are used, there is never a negative impact from defragmentation and no scheduling is ever required. Best of all, fragmentation is never an issue again.

So while that phrase “it’s all taken care of” might require investigation after the fact in many other cases just to verify it, automatic defragmentation will allow you to cross defragmentation off that list forever.