Hi, I have tried to understand how CEPH stores and retrieves data, and I have a few beginners questions about this explanation http://ceph.com/wp-content/uploads/2012/12/pg-placement1.png 1. hash("foo"); what exactly is foo, is that the filename that the client tries to write, or is it the object <VM-UUID>.<INC COUNTER> file in the pool? 2. What hash() algorithm is used in this implementation, is there any risk for collisions? 3. The calculation of both the PG and Target OSD occurs on the client using CRUSH after retrieving a cluster map for OSD,MON-map? Please advise. thx will _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |