* Re: hitting 2048 threads/jobs limit
[not found] ` <CANvN+ekk7A=ZQsgkeS0RWNVaJ7RfX8ALc7+UagobdhgUp9zqJw@mail.gmail.com>
@ 2017-01-23 14:17 ` Tobias Oberstein
0 siblings, 0 replies; 2+ messages in thread
From: Tobias Oberstein @ 2017-01-23 14:17 UTC (permalink / raw)
To: Andrey Kuzmin, fio
Hi Andrey,
thanks, got it .. FIO_MAX_JOB was bumped to 4096 like 4 months ago in
this commit
https://github.com/axboe/fio/commit/cab2472e2e9e84877e65c6aa68b86899956c8a28
So I just needed to rebuild from source.
Now I can max out this box using 2800 threads at 4.7 mio 8kB random read
IOPS =)
Awesome.
Cheers,
/Tobias
oberstet@svr-psql19:~/scm/parcit/RA/adr/system/docs$ sudo
/opt/fio/bin/fio postgresql_storage_workload.fio
randread: (g=0): rw=randread, bs=8192B-8192B,8192B-8192B,8192B-8192B,
ioengine=sync, iodepth=1
...
fio-2.17-17-g9cf1
Starting 2800 threads
Jobs: 2736 (f=37103):
[f(7),_(1),f(3),_(1),f(3),_(2),f(1),_(1),f(1),_(14),f(1),_(2),f(3),_(4),f(1),_(1),f(1),_(3),f(2),_(11),f(1),_(1),f(5),_(1),f(17),_(1),f(3),_(1),f(56),_(1),f(10),_(1),f(4),_(1),f(5),_(1),f(6),_(1),f(489),_(1),f(9),_(1),f(191),_(1),f(139),_(1),f(22),_(1),f(229),_(1),f(19),_(1),f(28),_(1),f(138),_(1),f(480),_(1),f(6),_(1),f(9),_(1),f(2),_(1),f(284),_(1),f(338),_(1),f(223)][100.0%][r=32.8GiB/s,w=0KiB/s][r=4297k,w=0
IOPS][eta 00m:00s]
randread: (groupid=0, jobs=2800): err= 0: pid=99065: Mon Jan 23 15:16:24
2017
read: IOPS=4732k, BW=36.2GiB/s (38.8GB/s)(4332GiB/120007msec)
clat (usec): min=43, max=26678, avg=583.62, stdev=572.89
lat (usec): min=43, max=26678, avg=583.70, stdev=572.89
clat percentiles (usec):
| 1.00th=[ 153], 5.00th=[ 185], 10.00th=[ 209], 20.00th=[ 253],
| 30.00th=[ 302], 40.00th=[ 354], 50.00th=[ 418], 60.00th=[ 502],
| 70.00th=[ 612], 80.00th=[ 772], 90.00th=[ 1064], 95.00th=[ 1448],
| 99.00th=[ 3152], 99.50th=[ 3984], 99.90th=[ 5856], 99.95th=[ 6688],
| 99.99th=[ 8768]
lat (usec) : 50=0.01%, 100=0.01%, 250=19.14%, 500=40.68%, 750=19.20%
lat (usec) : 1000=9.54%
lat (msec) : 2=8.76%, 4=2.19%, 10=0.48%, 20=0.01%, 50=0.01%
cpu : usr=0.45%, sys=3.68%, ctx=568493883, majf=0, minf=5598
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=567819066,0,0, 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=36.2GiB/s (38.8GB/s), 36.2GiB/s-36.2GiB/s
(38.8GB/s-38.8GB/s), io=4332GiB (4652GB), run=120007-120007msec
Disk stats (read/write):
nvme0n1: ios=35440924/0, merge=0/0, ticks=18474800/0,
in_queue=18776444, util=99.79%
nvme1n1: ios=35489875/0, merge=0/0, ticks=16143360/0,
in_queue=16950032, util=100.00%
nvme2n1: ios=35489702/0, merge=0/0, ticks=27449436/0,
in_queue=28772208, util=100.00%
nvme3n1: ios=35489494/0, merge=0/0, ticks=19419180/0,
in_queue=19731140, util=99.71%
nvme4n1: ios=35489290/0, merge=0/0, ticks=18055916/0,
in_queue=19064904, util=100.00%
nvme5n1: ios=35489126/0, merge=0/0, ticks=23692336/0,
in_queue=24074852, util=99.79%
nvme6n1: ios=35488938/0, merge=0/0, ticks=19653808/0,
in_queue=19951296, util=99.67%
nvme7n1: ios=35488789/0, merge=0/0, ticks=17548548/0,
in_queue=17833728, util=99.70%
nvme8n1: ios=35488597/0, merge=0/0, ticks=25328664/0,
in_queue=25684916, util=99.69%
nvme9n1: ios=35488409/0, merge=0/0, ticks=15241580/0,
in_queue=15956076, util=100.00%
nvme10n1: ios=35488240/0, merge=0/0, ticks=16489904/0,
in_queue=16750844, util=99.82%
nvme11n1: ios=35488059/0, merge=0/0, ticks=25966172/0,
in_queue=27363044, util=100.00%
nvme12n1: ios=35487878/0, merge=0/0, ticks=16283216/0,
in_queue=16521860, util=99.78%
nvme13n1: ios=35487713/0, merge=0/0, ticks=17994988/0,
in_queue=18756864, util=100.00%
nvme14n1: ios=35487556/0, merge=0/0, ticks=16564664/0,
in_queue=16794600, util=99.90%
nvme15n1: ios=35487377/0, merge=0/0, ticks=14727728/0,
in_queue=14983204, util=99.85%
oberstet@svr-psql19:~/scm/parcit/RA/adr/system/docs$
^ permalink raw reply [flat|nested] 2+ messages in thread