IT directors and system administrators have a tendency to be a bit busy (understatement of the century). They are normally so overwhelmed with the day-to-day business of keeping systems up and running and putting out fires as well as implementing the latest company priorities, that often a subject as “basic” as file fragmentation gets pushed to the bottom of the list. But interestingly, if they were able to take a closer look, they’d find that hard drive fragmentation is more of an insidious disease–and having much more of an impact on their systems–than they might think.
First, because a disk is mechanical and is the slowest element in a system, it is the “weakest link.” Hence, anything which slows it down is not only slowing the system’s slowest element, but overall system performance as well. Defrag has long since been proven to improve system performance dramatically. While fragmentation often causes performance slows and in severe cases impact system reliability. This slowing is caused by the extra I/O traffic involved to retrieve fragmented files, and severe fragmentation not only causes additional wear-and-tear on a disk drive but can also cause system hangs and crashes.
The effect on performance is felt all over a site through slow application response. If fragmentation has been an issue for some time, users might begrudgingly take slow performance as a given, and the system administrator or IT director might never hear about it and might therefore never react to it. But a whole company having to wait for files and applications means that production is occurring more slowly, which is impacting the company’s bottom line and everyone’s paychecks and job stability. This “basic” problem is creating that much of an effect!
This is especially true in a scenario which is not all that uncommon: trying to make hardware that is slightly outdated continue to perform. Disks are full and systems are stretched to their limits with utilization of resources such as memory. In such cases, fragmentation is having a tremendous impact.
Many IT personnel do know something about fragmentation and that it needs to be addressed, and have dealt with it by utilizing scheduled defragmentation solutions. The defragmenter runs periodically on specified drives at specified times, and the datacenter personnel consider that issue taken care of and move onto other things. But here’s an important question: how many of those “other things” might not be issues if fragmentation were fully addressed? Scheduled fragmentation is now an outdated solution, leaving performance-crippling fragmentation behind in between runs, and in some cases not improving performance at all.
A true “acid test” for the impact of fragmentation is to try a modern defrag solution, one which is fully automatic, has no impact on system performance when running, and operates transparently in the background whenever system resources are available. An additional benefit is no more valuable IT hours needed to schedule defragmentation.
Such a test will reveal the truth: that fragmentation is a far more insidious disease than one might think, and ridding a site of it maximizes performance–permanently.
