Wir haben einen RS 2000 SAS G8 und im Rettungssystem mit dem VIRTIO Treiber komme ich auf folgendes Ergebnis:
Code
root@xxx /mnt/home # fio --name=test --ioengine=libaio --rw=randrw --bs=4k --size=10G --numjobs=4 --runtime=900 --direct=1 --group_reporting --filename=/mnt/home/fio-testfile --log_avg_msec=1000 --write_iops_log=fio_iops --write_bw_log=fio_bw
test: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=1
...
fio-3.32
Starting 4 processes
test: Laying out IO file (1 file / 10240MiB)
Jobs: 4 (f=4): [m(4)][100.0%][r=612KiB/s,w=664KiB/s][r=153,w=166 IOPS][eta 00m:00s]
test: (groupid=0, jobs=4): err= 0: pid=2810: Thu Feb 13 09:00:32 2025
read: IOPS=245, BW=982KiB/s (1005kB/s)(863MiB/900008msec)
slat (usec): min=5, max=11743, avg=27.13, stdev=53.52
clat (usec): min=6, max=5588.8k, avg=16015.81, stdev=33464.60
lat (usec): min=104, max=5588.8k, avg=16042.94, stdev=33464.94
clat percentiles (usec):
| 1.00th=[ 245], 5.00th=[ 3392], 10.00th=[ 4686], 20.00th=[ 6128],
| 30.00th=[ 7242], 40.00th=[ 8356], 50.00th=[ 9503], 60.00th=[ 10814],
| 70.00th=[ 13173], 80.00th=[ 17695], 90.00th=[ 30540], 95.00th=[ 50070],
| 99.00th=[113771], 99.50th=[166724], 99.90th=[337642], 99.95th=[438305],
| 99.99th=[658506]
bw ( KiB/s): min= 36, max= 1740, per=100.00%, avg=987.59, stdev=70.75, samples=3575
iops : min= 9, max= 435, avg=246.72, stdev=17.67, samples=3575
write: IOPS=245, BW=982KiB/s (1006kB/s)(863MiB/900008msec); 0 zone resets
slat (usec): min=5, max=10748, avg=28.24, stdev=55.57
clat (usec): min=4, max=5551.0k, avg=218.18, stdev=11822.31
lat (usec): min=90, max=5551.0k, avg=246.42, stdev=11822.51
clat percentiles (usec):
| 1.00th=[ 114], 5.00th=[ 127], 10.00th=[ 135], 20.00th=[ 145],
| 30.00th=[ 153], 40.00th=[ 161], 50.00th=[ 169], 60.00th=[ 180],
| 70.00th=[ 190], 80.00th=[ 204], 90.00th=[ 233], 95.00th=[ 269],
| 99.00th=[ 412], 99.50th=[ 685], 99.90th=[ 1909], 99.95th=[ 3523],
| 99.99th=[28443]
bw ( KiB/s): min= 28, max= 2071, per=100.00%, avg=988.94, stdev=82.10, samples=3572
iops : min= 7, max= 517, avg=247.05, stdev=20.51, samples=3572
lat (usec) : 10=0.01%, 20=0.01%, 50=0.01%, 100=0.07%, 250=46.98%
lat (usec) : 500=3.24%, 750=0.16%, 1000=0.09%
lat (msec) : 2=0.65%, 4=2.25%, 10=23.63%, 20=14.41%, 50=6.01%
lat (msec) : 100=1.86%, 250=0.52%, 500=0.09%, 750=0.01%, 1000=0.01%
lat (msec) : 2000=0.01%, >=2000=0.01%
cpu : usr=0.19%, sys=0.49%, ctx=442374, majf=0, minf=63
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=220855,220984,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=982KiB/s (1005kB/s), 982KiB/s-982KiB/s (1005kB/s-1005kB/s), io=863MiB (905MB), run=900008-900008msec
WRITE: bw=982KiB/s (1006kB/s), 982KiB/s-982KiB/s (1006kB/s-1006kB/s), io=863MiB (905MB), run=900008-900008msec
Disk stats (read/write):
vda: ios=220826/221317, merge=44/179, ticks=3533279/49207, in_queue=3582557, util=100.00%
fio --name=test --ioengine=libaio --rw=randrw --bs=4k --size=10G --numjobs=4 13.16s user 50.23s system 6% cpu 15:43.43 total
Alles anzeigen
0.98 MiB/s sind definitiv zu wenig! Hab schon mit dem Support gesprochen, der uns dann auch verschoben hat. Die Geschwindigkeit wurde minimal besser. Was meint ihr?