Bill Lancaster writes:
>We did [condenses] because [they] gained us significant
>performance advantage for
>primarily two reasons.  The first reason is that it allowed us to very
>nearly condense all of the in-use space to the outer edges of the disk
>platters.  The second reason was that it allowed us to have larger chunks of
>free space to use. ...Let's say that
>we have a 50 percent full disk but the data is spread out all over the disk.
>Let's agree that, for the sake of discussion, given the radius of the disk
>is "r", the average head travel distance is r/2.  If that is true, is it not
>equally true that if you condense your disk to where all of the data is on
>the out edges of the platters doesn't the average head travel distance drop
>to r/4?
 
Yes, but the average SEEK time doesn't drop by anything like that ratio.
The reason is that on all modern disc drives, seek time is not a linear
function of seek distance. In fact, on a drive where a track-to-track seek
takes 5ms, a full-disc seek may take only 20ms. The reason is that the time
for short seeks is dominated by some constant-time or nearly constant-time
components: command processing, acceleration and deceleration, settling
time and servo reacquisition. So while the distance traveled after a
condense may decrease by 50%, the seek time may decrease by less than 25%.
 
>I would like to hear from any other lurkers on this list what kind of
>experience they have had in defragging/condensing etc. their systems both
>pre-5.0 and later.  Regardless of what product they use.
 
A client of mine used a disc third-party disc defragger that gave excellent
results. It attained this result by damaging the file system and forcing a
reload, which, though it did accomplish the defragmentation, seemed like
kind of a roundabout way of doing it. (To be fair, the client was using the
defrag utility in a way that neither its designers nor I had anticipated.
Others have used the same product quite successfully in less demanding --
and more reasonable -- ways.)
 
I think it's a little bit disingenuous of HP to say that VOLUTIL CONTIGVOL
isn't a disc defrag utility. Granted, it doesn't rearrange file extents
like the third-party utilities do. However, it still condenses free space
and therefore reduces file fragmentation when files are later created or
expanded. It most certainly produces the effect that Bill Lancaster
describes above.
 
-- Bruce
 
---------------------------------------------------------------------------
Bruce Toback    Tel: (602) 996-8601| My candle burns at both ends;
OPT, Inc.            (800) 858-4507| It will not last the night;
11801 N. Tatum Blvd. Ste. 142      | But ah, my foes, and oh, my friends -
Phoenix AZ 85028                   | It gives a lovely light.
[log in to unmask]                 |     -- Edna St. Vincent Millay