From: bugzilla-daemon@bugzilla.kernel.org
To: linux-ext4@vger.kernel.org
Subject: [Bug 78651] Write performance of ext4 degrades linearly as volume fills
Date: Tue, 08 Jul 2014 13:42:19 +0000 [thread overview]
Message-ID: <bug-78651-13602-2MPQ03SQUo@https.bugzilla.kernel.org/> (raw)
In-Reply-To: <bug-78651-13602@https.bugzilla.kernel.org/>
https://bugzilla.kernel.org/show_bug.cgi?id=78651
--- Comment #9 from Joseph D. Wagner <joe@josephdwagner.info> ---
This is entirely unexpected. Performance wasn't that bad. The only thing I
did differently was using this option in mke2fs:
"-E lazy_itable_init=0,lazy_journal_init=0"
Could the lazy initialization be tripping it up? That doesn't make much sense,
because I had been having this problem for some time before doing all these
formatting experiments.
I will attach ftrace and memstats shortly.
Joseph D. Wagner
####
#### Exhibit G
####
#### EXT4 256 MB journal
####
# su -c 'cd /root; nice ionice -c3 /usr/local/bin/lvbackup /dev/data/home
/mnt/backup'
Password:
/sys/kernel/debug/tracing/events/jbd2/jbd2_checkpoint/filter
/sys/kernel/debug/tracing/events/jbd2/jbd2_run_stats/filter
DUMP: mount: bad UUID
DUMP: Date of this level 0 dump: Mon Jul 7 22:32:53 2014
DUMP: Dumping /dev/data/backup-data-home (an unlisted file system) to
/mnt/backup/backup-data-home-20140707.223252-0.img
DUMP: Label: none
DUMP: Writing 64 Kilobyte records
DUMP: mapping (Pass I) [regular files]
DUMP: mapping (Pass II) [directories]
DUMP: estimated 1664540695 blocks.
DUMP: writing QFA positions to
/mnt/backup/backup-data-home-20140707.223252-0.qfa
DUMP: Volume 1 started with block 1 at: Mon Jul 7 22:33:10 2014
DUMP: dumping (Pass III) [directories]
DUMP: dumping (Pass IV) [regular files]
DUMP: 1.52% done at 84065 kB/s, finished in 5:25
DUMP: 3.38% done at 93765 kB/s, finished in 4:45
DUMP: 5.21% done at 95705 kB/s, finished in 4:34
DUMP: 7.13% done at 98469 kB/s, finished in 4:21
DUMP: 9.16% done at 101232 kB/s, finished in 4:08
DUMP: 11.09% done at 102248 kB/s, finished in 4:01
DUMP: 13.24% done at 104623 kB/s, finished in 3:50
DUMP: 15.25% done at 105362 kB/s, finished in 3:43
DUMP: 17.08% done at 104930 kB/s, finished in 3:39
DUMP: 19.05% done at 105326 kB/s, finished in 3:33
DUMP: 21.08% done at 106019 kB/s, finished in 3:26
DUMP: 22.80% done at 105141 kB/s, finished in 3:23
DUMP: 24.57% done at 104591 kB/s, finished in 3:20
DUMP: 26.17% done at 103451 kB/s, finished in 3:18
DUMP: 28.00% done at 103348 kB/s, finished in 3:13
DUMP: 29.65% done at 102589 kB/s, finished in 3:10
DUMP: 31.33% done at 102054 kB/s, finished in 3:06
DUMP: 32.94% done at 101356 kB/s, finished in 3:03
DUMP: 34.72% done at 101206 kB/s, finished in 2:58
DUMP: 36.69% done at 101617 kB/s, finished in 2:52
DUMP: 38.28% done at 100972 kB/s, finished in 2:49
DUMP: 39.92% done at 100523 kB/s, finished in 2:45
DUMP: 41.82% done at 100747 kB/s, finished in 2:40
DUMP: 43.61% done at 100688 kB/s, finished in 2:35
DUMP: 45.59% done at 101036 kB/s, finished in 2:29
DUMP: 47.33% done at 100878 kB/s, finished in 2:24
DUMP: 48.89% done at 100304 kB/s, finished in 2:21
DUMP: 50.36% done at 99618 kB/s, finished in 2:18
DUMP: 51.93% done at 99191 kB/s, finished in 2:14
DUMP: 53.74% done at 99175 kB/s, finished in 2:09
DUMP: 55.55% done at 99220 kB/s, finished in 2:04
DUMP: 57.41% done at 99354 kB/s, finished in 1:58
DUMP: 59.29% done at 99500 kB/s, finished in 1:53
DUMP: 61.40% done at 100007 kB/s, finished in 1:47
DUMP: 62.90% done at 99529 kB/s, finished in 1:43
DUMP: 64.38% done at 98954 kB/s, finished in 1:39
DUMP: 65.86% done at 98497 kB/s, finished in 1:36
DUMP: 67.41% done at 98173 kB/s, finished in 1:32
DUMP: 68.88% done at 97749 kB/s, finished in 1:28
DUMP: 70.55% done at 97621 kB/s, finished in 1:23
DUMP: 72.06% done at 97286 kB/s, finished in 1:19
DUMP: 73.46% done at 96809 kB/s, finished in 1:16
DUMP: 75.17% done at 96766 kB/s, finished in 1:11
DUMP: 76.74% done at 96552 kB/s, finished in 1:06
DUMP: 78.29% done at 96316 kB/s, finished in 1:02
DUMP: 79.98% done at 96264 kB/s, finished in 0:57
DUMP: 81.59% done at 96111 kB/s, finished in 0:53
DUMP: 83.19% done at 95963 kB/s, finished in 0:48
DUMP: 84.59% done at 95591 kB/s, finished in 0:44
DUMP: 85.90% done at 95129 kB/s, finished in 0:41
DUMP: 87.31% done at 94801 kB/s, finished in 0:37
DUMP: 88.88% done at 94635 kB/s, finished in 0:32
DUMP: 90.40% done at 94436 kB/s, finished in 0:28
DUMP: 92.01% done at 94342 kB/s, finished in 0:23
DUMP: 93.66% done at 94287 kB/s, finished in 0:18
DUMP: 95.48% done at 94409 kB/s, finished in 0:13
DUMP: 97.04% done at 94271 kB/s, finished in 0:08
DUMP: 98.43% done at 93972 kB/s, finished in 0:04
DUMP: 99.76% done at 93637 kB/s, finished in 0:00
DUMP: Closing /mnt/backup/backup-data-home-20140707.223252-0.img
DUMP: Volume 1 completed at: Tue Jul 8 03:29:43 2014
DUMP: Volume 1 1664675072 blocks (1625659.25MB)
DUMP: Volume 1 took 4:56:33
DUMP: Volume 1 transfer rate: 93557 kB/s
DUMP: 1664675072 blocks (1625659.25MB) on 1 volume(s)
DUMP: finished in 17783 seconds, throughput 93610 kBytes/sec
DUMP: Date of this level 0 dump: Mon Jul 7 22:32:53 2014
DUMP: Date this dump completed: Tue Jul 8 03:29:43 2014
DUMP: Average transfer rate: 93557 kB/s
DUMP: Archiving dump to /mnt/backup/backup-data-home-20140707.223252-0.toc
DUMP: DUMP IS DONE
Kernel buffer statistics:
Note: "entries" are the entries left in the kernel ring buffer and are not
recorded in the trace data. They should all be zero.
CPU: 0
entries: 0
overrun: 0
commit overrun: 0
bytes: 1400
oldest event ts: 24220.806656
now ts: 25076.920627
dropped events: 0
read events: 614
CPU: 1
entries: 0
overrun: 0
commit overrun: 0
bytes: 3168
oldest event ts: 23960.588658
now ts: 25076.920710
dropped events: 0
read events: 756
CPU: 2
entries: 0
overrun: 0
commit overrun: 0
bytes: 1092
oldest event ts: 24801.889841
now ts: 25076.920772
dropped events: 0
read events: 865
CPU: 3
entries: 0
overrun: 0
commit overrun: 0
bytes: 1700
oldest event ts: 24551.112688
now ts: 25076.920828
dropped events: 0
read events: 702
CPU: 4
entries: 0
overrun: 0
commit overrun: 0
bytes: 400
oldest event ts: 24162.567661
now ts: 25076.920883
dropped events: 0
read events: 85
CPU: 5
entries: 0
overrun: 0
commit overrun: 0
bytes: 2168
oldest event ts: 22049.515769
now ts: 25076.920936
dropped events: 0
read events: 144
CPU: 6
entries: 0
overrun: 0
commit overrun: 0
bytes: 2868
oldest event ts: 21979.076019
now ts: 25076.920988
dropped events: 0
read events: 233
CPU: 7
entries: 0
overrun: 0
commit overrun: 0
bytes: 900
oldest event ts: 8080.390207
now ts: 25076.921039
dropped events: 0
read events: 9
CPU0 data recorded at offset=0x449000
65536 bytes in size
CPU1 data recorded at offset=0x459000
77824 bytes in size
CPU2 data recorded at offset=0x46c000
90112 bytes in size
CPU3 data recorded at offset=0x482000
73728 bytes in size
CPU4 data recorded at offset=0x494000
12288 bytes in size
CPU5 data recorded at offset=0x497000
16384 bytes in size
CPU6 data recorded at offset=0x49b000
24576 bytes in size
CPU7 data recorded at offset=0x4a1000
4096 bytes in size
Logical volume "backup-data-home" successfully removed
--
You are receiving this mail because:
You are watching the assignee of the bug.
next prev parent reply other threads:[~2014-07-08 13:42 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-06-22 5:44 [Bug 78651] New: Write performance of ext4 degrades linearly as volume fills bugzilla-daemon
2014-06-22 5:45 ` [Bug 78651] " bugzilla-daemon
2014-06-22 23:18 ` bugzilla-daemon
2014-07-05 17:14 ` bugzilla-daemon
2014-07-06 12:52 ` bugzilla-daemon
2014-07-06 17:57 ` bugzilla-daemon
2014-07-07 6:00 ` bugzilla-daemon
2014-07-07 6:53 ` bugzilla-daemon
2014-07-07 12:29 ` bugzilla-daemon
2014-07-08 13:42 ` bugzilla-daemon [this message]
2014-07-08 13:42 ` bugzilla-daemon
2014-07-08 13:43 ` bugzilla-daemon
2014-07-08 13:43 ` bugzilla-daemon
2014-07-08 13:43 ` bugzilla-daemon
2014-07-08 14:23 ` bugzilla-daemon
2014-07-09 3:27 ` bugzilla-daemon
2014-07-09 3:30 ` bugzilla-daemon
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=bug-78651-13602-2MPQ03SQUo@https.bugzilla.kernel.org/ \
--to=bugzilla-daemon@bugzilla.kernel.org \
--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).