On Mon, Jul 15, 2013 at 1:30 AM, Stefan Priebe - Profihost AG <s.priebe@xxxxxxxxxxxx> wrote: > Am 15.07.2013 10:19, schrieb Sylvain Munaut: >> Hi, >> >> I'm curious what would be the official recommendation for when you >> have multiple pools. >> In total we have 21 pools and that lead to around 12000 PGs for only 24 OSDs. >> >> The 'data' and 'metadata' pools are actually unused, and then we have >> 9 pools of 'rgw' meta data ( .rgw, .rgw.control, .users.uid, >> .users.email, .users, .log, .usage, .intent-log, .rgw.gc ). Then we >> have 2 pools of RBD volumes and 8 pools assigned to RGW buckets. We >> splitted those so we could control placement and replication level. >> (The OSDs are split into 'bulk' on SATA drives and 'fast' on SAS 10k >> drives). > > You have to devide the total pgs by the pools. Well, it's actually not quite that simple. Many of those pools have less data than others; when using multiple pools you want to account for the data balance between them as well. (And you can realistically put a whole lot more than 100 PGs on an OSD, so you should take advantage of that range). -Greg Software Engineer #42 @ http://inktank.com | http://ceph.com _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |