From: lokesh jaliminche <lokesh.jaliminche@gmail.com>
To: linux-ext4 <linux-ext4@vger.kernel.org>
Subject: improved performance in case of data journaling
Date: Wed, 2 Dec 2020 23:28:49 -0800 [thread overview]
Message-ID: <CAKJOkCoUGPctXEcJWZFo+d62CSBmYjxFr1D0j74OY2ijynMyUA@mail.gmail.com> (raw)
Hello everyone,
I have been doing experiments to analyze the impact of data journaling
on IO latencies. Theoretically, data journaling should show long
latencies as compared to metadata journaling. However, I observed that
when I enable data journaling I see improved performance. Is there any
specific optimization for data journaling in the write path?
fio Logs:
------------
metadata journaling enabled
========================================================================
Actual data written (calculated using iostat): 5820352 bytes
writer_2: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B,
(T) 4096B-4096B, ioengine=sync, iodepth=1
fio-3.16
Starting 1 process
writer_2: Laying out IO file (1 file / 102400MiB)
writer_2: (groupid=0, jobs=1): err= 0: pid=26021: Thu Dec 3 06:51:23 2020
write: IOPS=24.0k, BW=97.7MiB/s (102MB/s)(2930MiB/30001msec); 0 zone resets
clat (usec): min=12, max=3144, avg=16.63, stdev=22.33
lat (usec): min=12, max=3144, avg=16.70, stdev=22.33
clat percentiles (usec):
| 1.00th=[ 14], 5.00th=[ 15], 10.00th=[ 15], 20.00th=[ 15],
| 30.00th=[ 16], 40.00th=[ 16], 50.00th=[ 16], 60.00th=[ 16],
| 70.00th=[ 16], 80.00th=[ 17], 90.00th=[ 17], 95.00th=[ 18],
| 99.00th=[ 34], 99.50th=[ 44], 99.90th=[ 424], 99.95th=[ 562],
| 99.99th=[ 791]
bw ( KiB/s): min=99856, max=100168, per=100.00%, avg=99992.14,
stdev=99.10, samples=59
iops : min=24964, max=25042, avg=24998.03, stdev=24.78, samples=59
lat (usec) : 20=96.72%, 50=2.95%, 100=0.16%, 250=0.02%, 500=0.08%
lat (usec) : 750=0.06%, 1000=0.01%
lat (msec) : 2=0.01%, 4=0.01%
cpu : usr=34.96%, sys=44.69%, ctx=750093, majf=0, minf=11
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=0,750001,0,0 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):
WRITE: bw=97.7MiB/s (102MB/s), 97.7MiB/s-97.7MiB/s
(102MB/s-102MB/s), io=2930MiB (3072MB), run=30001-30001msec
Disk stats (read/write):
nvme0n1: ios=0/757110, merge=0/23753, ticks=0/12769, in_queue=116, util=99.74%
-------------------------------------------------------------------------------------------
data journaling enabled
========================================================================
Actual data written (calculated using iostat): 10070880 bytes
writer_2: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B,
(T) 4096B-4096B, ioengine=sync, iodepth=1
fio-3.16
Starting 1 process
writer_2: Laying out IO file (1 file / 102400MiB)
writer_2: (groupid=0, jobs=1): err= 0: pid=26103: Thu Dec 3 06:52:15 2020
write: IOPS=24.0k, BW=97.7MiB/s (102MB/s)(2930MiB/30001msec); 0 zone resets
clat (usec): min=3, max=283709, avg=15.16, stdev=946.43
lat (usec): min=3, max=283709, avg=15.21, stdev=946.43
clat percentiles (usec):
| 1.00th=[ 6], 5.00th=[ 7], 10.00th=[ 7], 20.00th=[ 8],
| 30.00th=[ 8], 40.00th=[ 8], 50.00th=[ 9], 60.00th=[ 9],
| 70.00th=[ 9], 80.00th=[ 11], 90.00th=[ 24], 95.00th=[ 28],
| 99.00th=[ 34], 99.50th=[ 44], 99.90th=[ 81], 99.95th=[ 676],
| 99.99th=[ 947]
bw ( KiB/s): min=48488, max=151408, per=99.87%, avg=99861.02,
stdev=13105.09, samples=59
iops : min=12122, max=37852, avg=24965.25, stdev=3276.27, samples=59
lat (usec) : 4=0.02%, 10=79.65%, 20=5.92%, 50=14.15%, 100=0.17%
lat (usec) : 250=0.01%, 500=0.02%, 750=0.03%, 1000=0.03%
lat (msec) : 2=0.01%, 4=0.01%, 10=0.01%, 20=0.01%, 50=0.01%
lat (msec) : 250=0.01%, 500=0.01%
cpu : usr=61.73%, sys=22.41%, ctx=115437, majf=0, minf=11
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=0,750001,0,0 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):
WRITE: bw=97.7MiB/s (102MB/s), 97.7MiB/s-97.7MiB/s
(102MB/s-102MB/s), io=2930MiB (3072MB), run=30001-30001msec
Disk stats (read/write):
nvme0n1: ios=0/766273, merge=0/766195, ticks=0/941966,
in_queue=407464, util=95.92%
Thanks & Regards,
Lokesh
next reply other threads:[~2020-12-03 7:29 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-03 7:28 lokesh jaliminche [this message]
2020-12-03 8:20 ` improved performance in case of data journaling Martin Steigerwald
2020-12-03 9:07 ` lokesh jaliminche
2020-12-22 17:47 ` Jan Kara
2020-12-22 22:24 ` Andreas Dilger
2020-12-28 4:06 ` lokesh jaliminche
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAKJOkCoUGPctXEcJWZFo+d62CSBmYjxFr1D0j74OY2ijynMyUA@mail.gmail.com \
--to=lokesh.jaliminche@gmail.com \
--cc=linux-ext4@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).