On Sat, 10 Feb 2001 09:50:15 +0100, Bernie <[EMAIL PROTECTED]> wrote:
> IMHO you should point out that you didn't use readahead. A test with max
> readahead (and possibly cache size) on all of them would also be
> interesting - or atleast write what the max is for each.
So much work, so little time.
Yes, I will do a few runs with MAX size and MAX lookahead separately so
you can see the effect vs my mimimalist approach. <G>
> Clarence wrote::
>> C) The only time readahead improves things is when you are loading a LARGE
>> file _on a 100% defragged and compacted drive_ for the first time.
> Sorry that's incorrect. The gain is *very* - no make that *enormous* - if
> you have read-ahead and read only a small ammount of data from a file at a
> time in the program. I did a test with and without SmartDrive when reading
> a 10MB (IIRC, might have been more) file and reading 1 characther at a time
> from it (going to the extreme I know). I got tried of waiting after around
> 1 hour without the read-ahead. With read-ahead (using the max of 56K) it
> took less than 5 minutes. (And there we see the effect of the ammounts of
> calls).
Well SURE. To me, you are saying just what *I* said. That was a LARGE
file, you were reading it for the first time essentially, because you
were reading only a small amount (your readahead size) at a time, and
of course I don't know the extent of your fragmentation, but you must
agree that if your disk was badly fragmented most of the readahead data
is useless ?
I DO agree that when readahead CAN do any good, it does a LOT of good.
I just don't think it does any good MOST of the time, and on most HDs,
(fragmented) it does harm. ;-)
- Clarence Verge
- Still using Arachne V1.62 ....
