Re: Diskwarrior question (sort of)

Kyle Jones wrote:

> Have either of you done benchmarks, even rough ones, to see how
> much you gain (or lose) by defragging? I ask because modern
> filesystems are designed to layout files on the disk for close to
> optimal access and that pattern isn't necessarily in contiguous
> block order. It's possible that your defragmentation tools might
> be hurting rather than helping.

Kyle:

Quantitatively, the answer to your question is no, I haven't done any
benchmarking. Qualitatively, the answer is yes. I can tell when
read/write performance is getting sluggish, or when it's starting to
take longer to load a large directory or calculate all the file sizes
in a Finder window...and there are other performance indicators that
are specific to the kinds of apps I run and what I do with them. For
example, digital audio recording to disk creates large files that work
much better when the app doesn't have to hunt and peck all over the
disk for file fragments on playback. Similarly, there's a performance
hit when the app has to write to a fragmented disk.

If you're saying that compulsive defragging just for the sake of
defragging isn't always advisable, perhaps you're right. I don't have
any specialized knowledge about the ability of "modern filesystems" to
accommodate file and disk fragmentation, but I do know that over time,
my available disk space and files become fragmented. That part I HAVE
quantified -- I can see the number of file and disk fragments in
TechTool Pro 4, and there is a consistently reliable correlation
between high frag counts and decreased performance.

Your question sounds more theoretical than empirical, but maybe I'm
inferring more than I should. Do you have direct, first-hand knowledge
of cases in which defragging actually decreases performance?

.

Relevant Pages

  • Re: Disk Defrag in MCE environment
    ... > recording is fragged. ... Defragger recommends defragging the drive. ... > is absolutely no problem with 18 fragments in a one hour show. ... Certainly you are correct that defragging adds additional wear to the disk ...
    (microsoft.public.windows.mediacenter)
  • Re: Diskwarrior question (sort of)
    ... >> Have either of you done benchmarks, even rough ones, to see how ... digital audio recording to disk creates large files that work ... > hit when the app has to write to a fragmented disk. ... you saw tangible benefits or you just saw defragging as a "best ...
    (comp.sys.mac.apps)
  • Re: Is defraging necessary?
    ... thousands of transactions per minute where defragging may be of some ... In this case, with an average access time of 12msec, you'll spend over six minutes just moving the head around. ... An NTFS system will suck up the file with ONE head movement. ... In your example: 8000 fragments. ...
    (microsoft.public.windowsxp.general)
  • Re: Defrag strategy to extend disk life
    ... defragging or not defragging ... one way or the other is *not* going to affect disk wear in any ... it's always rotating...the armature's movement is not what is going to wear ... Whether this increased disk life or reduced noise, ...
    (microsoft.public.win2000.general)
  • Re: Disk Scanning & Defragmentation
    ... And of course fewer disks wear out due to excessive defragging. ... doubt the two camps here actually cover the reality of disk failure. ... out bearings to dust to a platter's head failure to head ...
    (microsoft.public.windowsxp.general)