Hello, I upgraded a staging ceph cluster from latest Firefly to latest Hammer last week. Everything went fine overall and I would like to share my observations so far: a. every OSD upgrade lasts appr. 3 minutes. I doubt there is any way to speed this up though b. rados bench with different block sizes and different number of threads produces consistently 15-20% better write/read IOPS/throughput compared to Firefly. Subsequently, CPU load on OSD nodes was lower during the bench c. OSD apply latency increased 2x-3x for all OSDs. No clue though why this is happening. Commitcycle/journal latencies are at the same level. You may notice the effect for the apply latency at the uploaded image [1] It would be nice if someone else also shares his/her experience after upgrading to Hammer or/and propose more core metrics that should be looked at after major version upgrades. [1] https://up1.ca/#V5vcso6i8IQ01Se62NJqng Regards, Kostis _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |