Intended setup (only focusing on OSD server here, of course monitor and
other servers will be part of the intended cluster):
* 1 single Cluster, spanning across 2 datacenters:
* 6 OSD servers (each containing 5 OSD’s/disks) in Datacenter1
* 6 OSD servers (each containing 5 OSD’s/disks) in Datacenter2

* Erasure Code level k=3, m=2 (total overhead for the entire cluster should
be: 40%)
* ISA-L Erasure Code optimization
* Our requirements regarding IOPS, latency and throughput are low
* Survive failure of 2 OSD disks
* Survive failure of 1 OSD server

Is there a way to keep Erasuere Codes for the 6 OSD servers located in DC1
at DC1, so we can prevent high network throughput between DC1 and DC2
during rebuild because an entire OSD server fails? If yes, how?

Best regards,
ceph-users mailing list

Reply via email to