we have a cluster with very low load, however, "ceph osd perf" shows high commit_latency and apply_latency. root@stor-mgt01:~# ceph -s cluster: id: 3d1ec789-829d-4e0f-b707-9363356a68f1 health: HEALTH_WARN application not enabled on 3 pool(s) services: mon: 3 daemons, quorum a,b,c mgr: a(active) mds: rook-ceph-filesystem-1/1/1 up {0=rook-ceph-filesystem-a=up:active}, 1 up:standby-replay osd: 55 osds: 55 up, 55 in rgw: 3 daemons active data: pools: 18 pools, 3328 pgs objects: 1.06M objects, 1.37TiB usage: 4.18TiB used, 141TiB / 145TiB avail pgs: 3328 active+clean io: client: 46.2KiB/s rd, 2.58MiB/s wr, 197op/s rd, 273op/s wr ---------------------------------------------------------------------------------------- osd commit_latency(ms) apply_latency(ms) 54 0 0 53 0 0 52 0 0 51 0 0 50 0 0 49 11 11 48 9 9 47 3 3 46 39 39 21 28 28 20 26 26 19 28 28 18 1 1 17 17 17 16 6 6 15 11 11 14 9 9 13 8 8 12 12 12 11 14 14 10 21 21 0 10 10 1 12 12 2 30 30 3 3 3 4 3 3 5 23 23 6 56 56 7 2 2 8 36 36 9 5 5 22 26 26 23 3 3 24 2 2 25 3 3 26 19 19 27 3 3 28 4 4 29 8 8 30 16 16 31 21 21 32 13 13 33 7 7 34 10 10 35 12 12 36 5 5 37 5 5 38 4 4 39 29 29 40 26 26 41 35 35 42 5 5 43 4 4 44 8 8 45 6 6