raidtest for zfs

[Pawel Jakub Dawidek]

On Sun, Sep 23, 2007 at 11:35:15PM +0200, Ivan Voras wrote:
> Pawel Jakub Dawidek wrote:
> 
> > ZFS ZVOL doesn't present raw blocks to use, like graid3/graid5, so it's
> > quite possible that because of how the data was written and how ZFS
> > prefetches the data it's faster than one disk.
> 
> So does ZVOL actually offer "cooked" data to its users (caching, lazy
> allocation?), similar to what would happen if the entire ZVOL was one
> big file in a "normal" ZFS file system?

Yes.

> Slightly off-topic: for most file systems, it used to be recommended
> that "lots of smallish files" be distributed in lots of directories in a
> radix-like fashion instead of keeping all files in one huge directory.
> Do you know if ZFS does something to help this situation, or are the old
> rules of thumb still applicable?

ZFS uses extendable hash data structure for directory entires, but you
would need to test it on your own to see if its fast enough for your
needs.

-- 
Pawel Jakub Dawidek                       http://www.wheel.pl
pjd at FreeBSD.org                           http://www.FreeBSD.org
FreeBSD committer                         Am I Evil? Yes, I Am!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-geom/attachments/20070925/3cd8fd5a/attachment.pgp