Heute habe ich einmal auf meinen RS 8000 dank Snapshots einen IO-Test mit UEFI und einmal ohne UEFI mit folgendem Befehl durchgeführt:
Code
fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=test3 --bs=4k --iodepth=64 --size=1G --readwrite=randrw
Das Gute an diesen Befehl ist, dass dieser sehr Ressourcenschonend ist und den Server nicht für etliche Minuten bis Stunden unter Beschlag nimmt.
Mit UEFI:
Code
test: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
fio-3.19
Starting 1 process
test: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=70.4MiB/s,w=70.4MiB/s][r=18.0k,w=18.0k IOPS][eta 00m:00s]
test: (groupid=0, jobs=1): err= 0: pid=5779: Mon May 30 18:41:28 2022
read: IOPS=15.3k, BW=59.9MiB/s (62.8MB/s)(512MiB/8544msec)
bw ( KiB/s): min=38888, max=76320, per=99.90%, avg=61283.94, stdev=13107.39, samples=16
iops : min= 9722, max=19080, avg=15320.94, stdev=3276.92, samples=16
write: IOPS=15.3k, BW=59.9MiB/s (62.9MB/s)(512MiB/8544msec); 0 zone resets
bw ( KiB/s): min=39008, max=76616, per=99.71%, avg=61197.69, stdev=13130.07, samples=16
iops : min= 9752, max=19154, avg=15299.38, stdev=3282.59, samples=16
cpu : usr=7.43%, sys=15.21%, ctx=131042, majf=0, minf=7
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
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=0.1%, >=64=0.0%
issued rwts: total=131040,131104,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=64
Run status group 0 (all jobs):
READ: bw=59.9MiB/s (62.8MB/s), 59.9MiB/s-59.9MiB/s (62.8MB/s-62.8MB/s), io=512MiB (537MB), run=8544-8544msec
WRITE: bw=59.9MiB/s (62.9MB/s), 59.9MiB/s-59.9MiB/s (62.9MB/s-62.9MB/s), io=512MiB (537MB), run=8544-8544msec
Disk stats (read/write):
vda: ios=128176/2, merge=0/1, ticks=6624/0, in_queue=6624, util=99.00%
Display More
Ohne UEFI:
Code
test: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
fio-3.19
Starting 1 process
test: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=25.1MiB/s,w=25.4MiB/s][r=6414,w=6505 IOPS][eta 00m:00s]
test: (groupid=0, jobs=1): err= 0: pid=1632: Mon May 30 18:47:27 2022
read: IOPS=6238, BW=24.4MiB/s (25.6MB/s)(512MiB/21004msec)
bw ( KiB/s): min= 5405, max=35816, per=100.00%, avg=24990.59, stdev=7048.41, samples=41
iops : min= 1351, max= 8954, avg=6247.59, stdev=1762.18, samples=41
write: IOPS=6241, BW=24.4MiB/s (25.6MB/s)(512MiB/21004msec); 0 zone resets
bw ( KiB/s): min= 5205, max=36136, per=100.00%, avg=24992.12, stdev=7140.86, samples=41
iops : min= 1301, max= 9034, avg=6248.00, stdev=1785.26, samples=41
cpu : usr=2.92%, sys=8.29%, ctx=131047, majf=0, minf=8
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
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=0.1%, >=64=0.0%
issued rwts: total=131040,131104,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=64
Run status group 0 (all jobs):
READ: bw=24.4MiB/s (25.6MB/s), 24.4MiB/s-24.4MiB/s (25.6MB/s-25.6MB/s), io=512MiB (537MB), run=21004-21004msec
WRITE: bw=24.4MiB/s (25.6MB/s), 24.4MiB/s-24.4MiB/s (25.6MB/s-25.6MB/s), io=512MiB (537MB), run=21004-21004msec
Disk stats (read/write):
vda: ios=130041/20895, merge=0/15, ticks=18667/23655, in_queue=42322, util=99.45%
Display More
Die Unterschiede in Bezug auf die IOPS sind wie Tag und Nacht.