When it came about, scheduled defragmentation was a lifesaver. Prior to its development, defragmentation had to be performed manually, which meant a human had to stay late at night or come in over the weekend to fire it off and make sure it ran. Scheduling defragmentation meant that the system administrator or other system personnel could set up a schedule for a disk or disks, then leave. When they returned in the morning the disks would be all neatly defragmented and performance would be restored.
When the PC revolution happened in the nineties, however, there were suddenly a lot more disks to deal with, and it was at that point that scheduled defragmentation began to get complicated. Instead of having to analyze one or two disks and set defragmentation schedules for them, it was suddenly twenty, fifty, or a hundred. Until networking technology had really evolved, each of the schedules had to be set computer-by-computer. But even when all schedules could be set remotely, it was still a time-consuming effort.
Now, of course, disks can and often do number in the thousands. To accurately analyze defragmentation on and schedule defragmentation across an entire site can be extremely taxing on an already-overworked IT staff.
As we more closely examine today’s computing environment, other problems arise. Business has gone global, which means access to drives happens well beyond the former “nine-to-five” limits. Additionally, many more staff telecommute, logging into a company’s system at all hours. When can defragmentation be scheduled so it won’t impact users? Such times have all but disappeared.
Unfortunately, fragmentation has anything but disappeared, and defragmentation is more required than ever. Because disks have grown so much larger along with file sizes, performance-crippling fragmentation builds up even faster than before. Because of this, scheduled defragmentation is no longer doing the job; fragmentation continues to build in between scheduled runs. And in some cases, as with very large disks, for example, fragmentation isn’t even being affected.
The answer to the burdensome scheduling complications and other problems associated with scheduled defragmentation is to simply do away with the “scheduled” part and adopt a completely automatic solution. Such a solution does not require scheduling as it only utilizes otherwise-idle resources. It does not impact any users on the system for the same reason. And best of all, it continuously addresses fragmentation so that systems always operate at maximum performance.