Stefan said:
Such an absolute statement is misleading. If a drive doesn't need
defragging, whether it's been a month or several since the last one, why
defrag it? And overly defragging puts extra wear and tear on the drive,
too. Plus, it's a waste of time.
Stef
Comes down to whether you're the type of person that prefers incremental
maintenance to keep things smooth or prefers to wait to do major or
catastrophic repairs before which you incurred less than stellar
operation. A little at a time or a lot all at once. You could pickup
each sock you drop in your bedroom and put it away or you could wait
several months and pick them all up to get rid of the huge mess.
You won't be saving on disk wear by doing the major defrag at infrequent
intervals versus doing minor defrags at frequent intervals. Do a lot
all at once but infrequently or a little each time and frequently.
Please explain how there is any difference in disk wear. If left, the
same little defrag you do frequently will be incorporated in the big
defrag you do once in awhile. What's defragged a little is still going
to be what pends until you get around to doing the big defrag. If your
hard disk is 10% fragmented, that's still going to be there if you wait
around to do a defrag when it gets up to 50% fragmented (one-fifth of
that 50% is the original 10% that you left around waiting until you do
the big defrag). You're not changing how much you defrag by waiting. A
fragmented file will take longer to load (if the fragments are in
different cylinders versus, say, a 10-fragment file with all fragments
in the same cylinder). A big defrag is going to include all those
little defrags you could've done earlier. So by waiting to do a big
defrag, you've lengthened over how long that file opened slower.
This argument has been used many times in the past and yet no one
presents any empirical study to prove a single huge defrag puts less
wear on the hard disk than a bunch of smaller defrags. It's become
urban legend or wives' tale proliferated by repetition without proof.
Many of those that do periodic defrags are scheduling them to run when
they are not at or using their computer, so what functionality is lost?
It's not like they care about the reduction in responsiveness of their
host during a little defrag because they're not at their computer to use
it. Tis easy to schedule a tiny defrag that typically takes less than
half an hour than wait for months to do a defrag and not be sure when it
will complete which means it could be still running when you return to
your host or interfere with other scheduled tasks.
Since Windows XP, automatic defragmentation has been part of Windows.
It's running even when you don't run your defragger. Of course, the
Windows-supplied auto-defragger has different goals than an on-demand
defragger but your hard disk is still getting defragged a little at a
time while Windows is running. In fact, because of this background
defragmentation, it is recommended you disable it if you use intend to
frequently defrag with a 3rd party tool. Since each defragger has
different algorithms for best placement of files, one will undo the
defrag by another.
With little defrags you minimize the impact on your host regarding
responsiveness: the impact is short-lived. The little defrag takes a
little time so it is easy to schedule and permit it sufficient time to
complete so it doesn't interfere with later scheduled jobs and it won't
be running when you return to your host. The little defragmentation
will still be there for a big infrequent defrag so you haven't saved
anything by waiting. With fragmented files, the heads are going to move
around MORE to find all those fragments which means there is MORE head
assembly wear for all those extra movements between the infrequent big
defrags. Alas, there is no good measuring tool to determine when the
wear from a little defrag will compensate for the wear from moving the
heads around to find all the fragments; if there were, all the
defraggers would include such a measurement to provide a guage by which
to measure when a defrag would REDUCE subsequent head wear.
The only way the argument wins for doing infrequent big defrags is to
not do them at all; i.e., push out the defrag interval so far that added
disk wear for ANY defragging is insignificant to the MTBF of the device.
That is, do frequent little defrags or infrequent big defrags OR don't
defrag at all.
Rather than regurgitate this repeated myth, perhaps someone can present
real empirical data (laboratory testing) showing overall head wear for
frequent small defrags along with reduced head movement to access
unfragmented files is more than the overall head wear for infrequent big
defrags along with increased head movement to access the fragmented
files between those big defrags. The little defrags are still
incorporated in the big defrags so you haven't saved on head wear there.
It's *between* the defrags to access unfragmented files were you save on
head wear. It's not whether little defrags cause more wear than big
defrags (which has yet to be empirically proven). It's about how much
you reduce the wear from accessing fragmented files (by making them
unfragmented) and when that reduction in file access wear exceeds the
wear generated by the defrags (a bunch of little ones or a few big
ones).
Do you know of a tool that records a history of how often the heads have
to move to different cylinders both during a defrag operation and also
outside of that defrag (to measure access of fragmented files)? Then
you might have some measure by which you could determine if all the head
movements for a defrag are less than the increased head movements to
access fragmented files. I've trialed a few defraggers but haven't seen
any that give you a decent measure of when you should defrag (and a
percentage level of fragmentation does not tell you the defrag's head
movements will be less than those for accessing the fragmented files).
You'd need to record how many head movements there were in a defrag.
Then monitor the head movements incurred ONLY to access the fragments of
a file thereafter. When the amount of head movement to access the
fragmented files got higher than the head movement incurred by a defrag
then you know you finally got to a ROI point in your hard disk
maintenance.
I prefer not to pay for