On 11/04/2013 08:01 PM, John R Pierce wrote: > On 11/4/2013 10:43 AM, Les Mikesell wrote: >> On Mon, Nov 4, 2013 at 12:15 PM, Markus Falb >>> <wnefal at gmail.com> wrote: >>>>>> >>> 3) NEVER let a zpool fill up above about 70% full, or the performance >>>>> really goes downhill. >>> >>>> Why is it? It sounds cost intensive, if not ridiculous. >>>> Disk space not to used, forbidden land... >>>> Is the remaining 30% used by some ZFS internals? >> Probably just simple physics. If ZFS is smart enough to allocate >> space 'near' other parts of the related files/directories/inodes it >> will have to do worse when there aren't any good choices and it has to >> fragment things into the only remaining spaces and make the disk heads >> seek all over the place. Might not be a big problem on SSD's though. > > even on 0 seek time SSDs, fragmenting files means more extents to track > and process in order to read that file. but why would this be much worse with ZFS than eg ext4?