In article <(E-Mail Removed)>,
(E-Mail Removed) says...
> Because 4k is the data size used when the system is "paging". It just seems
> to make the operating system a bit more "snappy" [in my estimation]. I would
> guess that it may eliminate extra overhead involved when using
> larger/smaller cluster sizes, and the system is making use of the pagefile.
I have a drive that is used to store small images, under 30k many times,
I have worked with the drive set at 512b and at the default 4k and even
larger - the 512b provides the best in unwasted slack space - and you
can really see this with 50,000+ files.
For database servers I move their data drive/array to larger cluster
sizes, 4k being way to small in my opinion.
Paging means little of you are not paging a lot.
What you have to do, to find the optimal size, is determine the size of
70% of your files and then determine the amount of wasted slack space
they consume and setup the cluster size for that. Sure, tracking small
cluster sizes is a performance hit, but wasted disk space is often more
of a problem for users.
--
(E-Mail Removed)
remove 999 in order to email me