Am 03.02.2016 um 17:24 schrieb Wade Holler: > AFAIK when using XFS, parallel write as you described is not enabled. Not sure I'm getting this. If I have multiple OSDs on the same NVMe (separated by different data-partitions) I have multiple parallel writes (one "stream" per OSD), or am I mistaken? > Regardless in a way though the NVMe drives are so fast it shouldn't > matter much the partitioned journal or other choice. Thanks, does anyone has benchmarks on this. How about the size of the journal? > What I would be more interested in is you replication size on the cache > pool. > > This might sound crazy but if your KVM instances are really that short > lived, could you get away with size=2 on the cache pool from and > availability perspective ? :) We are already on min_size=1, size=2 - we even ran for a while witz min_size=1, size=1, so we cannot squeeze out much more on that end. Greetings -Sascha- PS: Thanks a lot already for all the answers! _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |