Hi,
I am quite new to fio, and I cannot interpret bandwidth results, yet. I found some articles on the interpretation of fio results, but no answer for my specific question.
When I do *not* generate log files, the results seem clear to me. But with the "--write_bw_log" option I see a new "bw" section (which also reflects the numbers written to the logfile) and a discrepancy between these bandwidth numbers:
=========================================
root@knorkator:~/fio# ./fio/fio --numjobs=1 --iodepth=1 --direct=1 --runtime=13 --time_based --size=50M --directory=/opt/tmp --ioengine=libaio --rw=read --bs=4k --name=local-libaio-read-4k --write_bw_log=log/local-libaio-read-4k
local-libaio-read-4k: (g=0): rw=read, bs=4096B-4096B,4096B-4096B,4096B-4096B, ioengine=libaio, iodepth=1
fio-2.17-33-g3dd2e
Starting 1 process
Jobs: 1 (f=1): [R(1)][100.0%][r=91.2MiB/s,w=0KiB/s][r=23.6k,w=0 IOPS][eta 00m:00s]
local-libaio-read-4k: (groupid=0, jobs=1): err= 0: pid=13599: Fri Feb 3 17:26:41 2017
read: IOPS=23.5k, BW=91.5MiB/s (95.9MB/s)(1189MiB/13001msec)
slat (usec): min=5, max=127, avg= 7.56, stdev= 2.02
clat (usec): min=2, max=1058, avg=32.63, stdev= 8.33
lat (usec): min=33, max=1065, avg=40.19, stdev= 8.90
clat percentiles (usec):
| 1.00th=[ 29], 5.00th=[ 29], 10.00th=[ 29], 20.00th=[ 30],
| 30.00th=[ 30], 40.00th=[ 30], 50.00th=[ 30], 60.00th=[ 31],
| 70.00th=[ 32], 80.00th=[ 34], 90.00th=[ 39], 95.00th=[ 41],
| 99.00th=[ 58], 99.50th=[ 91], 99.90th=[ 126], 99.95th=[ 153],
| 99.99th=[ 213]
bw ( KiB/s): min= 3871, max=2048000, per=0.13%, avg=128690.23, stdev=16014.17
lat (usec) : 4=0.01%, 20=0.01%, 50=98.87%, 100=0.78%, 250=0.34%
lat (usec) : 500=0.01%, 750=0.01%
lat (msec) : 2=0.01%
cpu : usr=10.15%, sys=26.03%, ctx=304964, majf=0, minf=1802
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 rwt: total=304292,0,0, short=0,0,0, dropped=0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=91.5MiB/s (95.9MB/s), 91.5MiB/s-91.5MiB/s (95.9MB/s-95.9MB/s), io=1189MiB (1246MB), run=13001-13001msec
Disk stats (read/write):
sda: ios=301861/41, merge=0/25, ticks=8612/24, in_queue=8660, util=66.55%
=========================================
The bandwidth numbers on the "read: ..." line and close to the bottom claim that I measured a total average bandwidth of 91.5 MiB/s. That seems reasonable.
But there is also a "bw" section that claims an avg of 128690.23 KiB/s. What is the difference between these numbers?
When I modify iodepth or numjobs parameters, the discrepancy changes, e.g. with iodepth=8 the first number goes up to 278 MiB/s, while the "bw" number goes down to 42921.43 KiB/s.
Unfortunately, the "bw" numbers are written to the logfile, so that my plots do not show what I would expect.
Thanks in advance for any help,
-frank
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
