Pete Majka wrote:
> I have a raidz zpool comprised of 5 320GB SATA drives and I am seeing the
> following numbers.
>
> capacity operations bandwidth
> pool used avail read write read write
> ---------- ----- ----- ----- ----- ----- -----
> vol01 123G 1.33T 66 182 8.22M 9.85M
>
> The overall IO rate is very poor at the application level as the latency is
> high as well.
What are your goals?
> Can anyone point me in a good direction, as all the roads I have gone down
> have led to nowhere.
I can't tell if your workload is small, random reads. It does not
appear to be. However, if your workload is small, random reads then
see:
http://blogs.sun.com/relling/entry/zfs_raid_recommendations_space_performance
http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide
66 iops looks like the performance I would expect from a 3.5" 7,200 rpm disk.
However, there is no service time information from this output. You'd need to
check with iostat to see whether the disks are actually busy. If they are not
busy, then look elsewhere.
-- richard
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
