On 01/02/2017 08:43 AM, Lindsay Mathieson wrote:
Yes, but we have a requirement for frequent use of snapshots and rbd
snapshots are unusably slow. qcow2 on top of cephfs works reasonably well.
Can you say more about pool snapshot speed? Slow to make snapshot? Slow
to access snapshot? Slow to delete a snapshot? Slow to access original?
Slow to modify original? (Slow to modify at first or slow continually?)
What are circumstances where pool snapshots are appropriate? Is there a
way to speed them up?
Where does the slowdown come from? What exacerbates the slowness? What
minimizes it? Are there prospects for this to speed up in future Ceph
versions?
Thanks,
-kb, the Kent who was thinking of a simplistic way to maintain indexing
of data being incrementally archived, including occasionally--at points
when the index is self consistent--slicing off a snapshot for others to
read; that is, using snapshots for poor-man's tree locking where readers
don't have to honor no stinkin' advisory locks and the writer doesn't
have to laboriously maintain them, he just makes snapshots occasionally.
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
