On 2013-11-06 09:33, Sage Weil wrote:
On Wed, 6 Nov 2013, Loic Dachary wrote:
Hi Ceph,
People from Western Digital suggested ways to better take advantage
of
the disk error reporting... when one head out of ten fails :
disks can keep working with the nine remaining heads. Losing 1/10 of
the
disk is likely to result in a full re-install of the Ceph osd. But,
again, the disk could keep going after that, with 9/10 of its
original
capacity. And Ceph is good at handling osd failures.
Yeah...but if you lose 1/10 of a block device any existing local file
system is going to blow up. I suspet this is something that
newgangled
interfaces like Kinetic will be much better at....
I found some info on this at last in the SATA-IO 3.2 Spec, seemingly
it's more to with RAID rebuilds:
"Rebuild Assist: when a drive in a RAID configuration fails due to
excessive data errors, it is possible
to reconstruct the data from the failed drive from the remaining drives
– this is called a Rebuild. The
Rebuild Assist function speeds up the rebuild process by quickly
recognizing which data on the failed
drive is unreadable"
Source:
https://www.sata-io.org/sites/default/files/images/SATA-IO%20FAQ%20-%20071813a%20%283%29.pdf
There is also some interesting info on SSHDs:
"SSHD Optimization: a Solid State Hybrid Drive (SSHD) is an HDD that
contains some amount of Flash
memory, thus increasing the performance of the drive. The Hybrid
Information feature provides a
mechanism wherein a host can tell the drive which data to cache,
further enhancing the
performance of the SSHD. In today’s SATA drives, reading and writing
log data required the use of
non-queued commands, impacting overall system performance, especially
SSHDs. A new feature in
v3.2 allows such commands to be queued, minimizing the impact on
performance"
But it seems the manufacturers are taking a different path; the new
Seagate 3.5" Hybrid drives don't even support the ATA-8 NV Cache feature
set unfortunately, according to the product manual:
http://www.seagate.com/files/staticfiles/support/docs/manual/desktop%20sshd/100726566.pdf
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
