Ceph
I built this trying to get kubernetes working. I realized I couldn’t have dynamic state without finding a distributed filesystem.
ariadne@tiny-4:~$ sudo ceph status
cluster:
id: 06184e1d-d46d-43fb-80ef-8b485942ca80
health: HEALTH_OK
services:
mon: 3 daemons, quorum tiny-5,tiny-4,tiny-7 (age 12d)
mgr: tiny-5(active, since 2w), standbys: tiny-4, tiny-7
mds: 1/1 daemons up, 2 standby
osd: 4 osds: 4 up (since 13d), 4 in (since 7M)
data:
volumes: 1/1 healthy
pools: 4 pools, 97 pgs
objects: 7.92k objects, 30 GiB
usage: 88 GiB used, 2.4 TiB / 2.5 TiB avail
pgs: 97 active+clean
My first ceph cluster I tried with USB 3.0 thumbdrives … ceph eats bandwidth, so this is backed with 20G port channels with NVME drives.
ariadne@tiny-4:~$ sudo rados bench -p rbd 10 write --no-cleanup
hints = 1
Maintaining 16 concurrent writes of 4194304 bytes to objects of size 4194304 for up to 10 seconds or 0 objects
Object prefix: benchmark_data_tiny-4_4182550
sec Cur ops started finished avg MB/s cur MB/s last lat(s) avg lat(s)
0 0 0 0 0 0 - 0
1 16 222 206 823.898 824 0.138184 0.073974
2 16 438 422 843.911 864 0.12581 0.0742721
3 16 646 630 839.918 832 0.112153 0.075176
4 16 853 837 836.922 828 0.077436 0.0753808
5 16 1064 1048 838.323 844 0.0452554 0.0754338
6 16 1286 1270 846.589 888 0.0690823 0.075085
7 16 1489 1473 841.637 812 0.12503 0.0754491
8 16 1702 1686 842.924 852 0.0767748 0.0755216
9 16 1920 1904 846.146 872 0.0426611 0.0751712
10 16 2134 2118 847.123 856 0.0574978 0.0751677
Total time run: 10.0712
Total writes made: 2134
Write size: 4194304
Object size: 4194304
Bandwidth (MB/sec): 847.568
Stddev Bandwidth: 23.6868
Max bandwidth (MB/sec): 888
Min bandwidth (MB/sec): 812
Average IOPS: 211
Stddev IOPS: 5.92171
Max IOPS: 222
Min IOPS: 203
Average Latency(s): 0.075293
Stddev Latency(s): 0.0340845
Max latency(s): 0.221071
Min latency(s): 0.0187221