On 29.08.2019 16∶05 +0800, Wesley Peng wrote:
Hion 2019/8/29 15:50, Płaza Tomasz wrote:Is there anything to speed-up replication? Should I enable "rgw run syncthread" on all rgws not just zone endpoints?Did you check the network? A faster network connection should be helpful. Latency is ~4.8ms and bandwith is ~70Mb/s in total (sum in and out) between clusters. For me it looks like network is sufficient. Migrating bucket with 1.5TB video files took about an hour, but 5GB bucket with files 1-2kB in size took about 3 hours.
On average new added files are between 4 to 16 kB in size.
regards. |
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com