I recently purchased a root server, everything was nice and fine in the first couple days, however, right now my hosted service is nearly dying due to horrid disk performance.
I'm quite literally getting 1.94 MiB/s on disk writes. What's going on?
Code
sequential-write-10GiB-repeat: (g=0): rw=write, bs=(R) 1024KiB-1024KiB, (W) 1024KiB-1024KiB, (T) 1024KiB-1024KiB, ioengine=psync, iodepth=1
fio-3.42
Starting 1 process
sequential-write-10GiB-repeat: Laying out IO file (1 file / 10240MiB)
fio: terminating on signal 2
sequential-write-10GiB-repeat: (groupid=0, jobs=1): err= 0: pid=48025: Mon Sep 14 00:37:43 2026
write: IOPS=1, BW=1985KiB/s (2032kB/s)(236MiB/121757msec)
clat (usec): min=230, max=7524.8k, avg=475268.23, stdev=1575429.88
lat (usec): min=234, max=7524.8k, avg=475274.46, stdev=1575429.19
clat percentiles (usec):
| 1.00th=[ 237], 5.00th=[ 249], 10.00th=[ 253],
| 20.00th=[ 265], 30.00th=[ 273], 40.00th=[ 281],
| 50.00th=[ 289], 60.00th=[ 302], 70.00th=[ 322],
| 80.00th=[ 351], 90.00th=[ 445], 95.00th=[5469373],
| 99.00th=[6476006], 99.50th=[6543115], 99.90th=[7549748],
| 99.95th=[7549748], 99.99th=[7549748]
bw ( KiB/s): min=20480, max=55296, per=100.00%, avg=24064.00, stdev=7543.15, samples=20
iops : min= 20, max= 54, avg=23.50, stdev= 7.37, samples=20
lat (usec) : 250=6.36%, 500=85.17%
lat (msec) : >=2000=8.47%
cpu : usr=0.00%, sys=0.06%, ctx=69, majf=0, minf=0
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=0,236,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0.00ns, window=0.00ns, percentile=100.00%, depth=1
Run status group 0 (all jobs):
WRITE: bw=1985KiB/s (2032kB/s), 1985KiB/s-1985KiB/s (2032kB/s-2032kB/s), io=236MiB (247MB), run=121757-121757msecEdit: back to 1.86 GiB/s - that's some crazy fluctuation...