[ Thursday, September 30, 1999 ] Tom Livingston wrote:
> my tests seemed to indicate the same thing... 4k chunk was faster than 128k
>
> I understand that this doesn't necessarily fit in with the theory of these
> things.
I'm not so sure it doesn't, as when it comes down to it you need to keep
as many drives actively fetching data as possible (hence adjustable raid1
read-balance testing still on my to-do list), and with huge chunk sizes
you'll end up in large periods of time where you're only dealing with one
(or a small number of) drives.
All my (frequently scientific :) testing has shown quite the sweet spot
at 4k, even with my datafiles (none smaller than 1GB)
James