When it came about, scheduled defragmentation was a marvelous development. Prior to it, the only way to handle performance-crippling file fragmentation was manually–which usually meant staying late or coming on weekends so disks could be defragmented when users weren’t on the system. Scheduled defragmentation gave system administrators their nights and weekends back; defragmentation could then be scheduled to run when nobody was around, and both users and IT staff could show up the next morning and the drives were all defragmented and delivering their best performance.
But as we know all too well the only constant in life is change, and this has never been more true than it is in the computing universe. As a little invention called the Internet–which some unbelievably considered would be a passing fad or at best a novelty–began growing and being utilized in earnest for business, commerce which previously had operated during “standard business hours” was suddenly open to the entire world. If a company had a web site, that web site was accessible at any time. And if a company sold product over that web site, the capacity to take orders 24X7 was mandatory. Hence, the time window during which users weren’t accessing a system began disappearing.
As the web became more reliable and secure, companies began utilizing it for customer service, tech support, and other business activities. CRM and database applications resided on servers which were now constantly being accessed–and that time window became even smaller. Now in many places the web is replacing functionality which was once the province of wide area networks (WANs)–the interconnection and synergy between corporate branches and even different enterprises. It is not uncommon for a division in Hong Kong to have permanent access via the web to the home office in Atlanta, and vice versa. The result: for many servers, the time window for scheduled defragmentation during which users aren’t accessing the system has disappeared altogether.
IT departments have had to cope as best they could, and schedule defragmentation to run anyway. Yes, it impacts the system while it is running and users complain. But the alternative is not to defragment at all, and although users may not understand all the technical ramifications, that is a far worse alternative for performance.
The time has never been more appropriate for a defragmentation solution that operates and maintains performance just as many sites do–24X7. Such a solution needs to not impact performance in any way whatsoever, only utilizing idle resources and defragmenting whenever possible. Additionally, it should not absorb increasingly-valuable IT time with having to schedule it. Only in such a way can disk performance and reliability be constantly maximized–because now, that window to schedule defragmentation is nonexistent.
It is of vital benefit to system administrators and IT staffs everywhere to seek out and implement such solutions.
