All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tomasz Chmielewski <mangoo@wpkg.org>
To: LVM general discussion and development <linux-lvm@redhat.com>
Subject: [linux-lvm] why LVM reads when I tell it to write?
Date: Mon, 23 Mar 2009 16:38:09 +0100	[thread overview]
Message-ID: <49C7ACE1.4070000@wpkg.org> (raw)

Why is LVM reading large amounts of data when I tell it to write only?


For example, this is what iostat shows when writing to a raw partition
(when doing "dd if=/dev/zero of=/dev/sdb5", iostat output every second):

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              80.81         0.00      5171.72          0       5120

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              78.00         0.00      4992.00          0       4992

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              81.82         0.00      5236.36          0       5184

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              77.00         0.00      4928.00          0       4928

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              79.21         0.00      5069.31          0       5120


This is what happens if we go through LVM - look how much is read before 
it actually writes anything:

(when doing "dd if=/dev/zero of=/dev/test/test_volume").

What's interesting, if I add i.e. bs=64k to dd, I don't see these unexpected
reads on LVM any more. Why does it happen?


Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             212.87       851.49         0.00        860          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             956.57      3826.26         0.00       3788          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1083.84      4335.35         0.00       4292          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1016.00      4064.00         0.00       4064          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1035.64      4142.57         0.00       4184          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1034.00      4136.00         0.00       4136          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             965.66      3862.63         0.00       3824          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1025.00      4100.00         0.00       4100          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             991.92      3967.68         0.00       3928          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1032.00      4128.00         0.00       4128          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1009.90      4039.60         0.00       4080          0

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             366.33      1330.61      4040.82       1304       3960

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb            1016.16      4060.61        76.77       4020         76

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             383.17      1394.06      4055.45       1408       4096

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             741.00      2904.00      1800.00       2904       1800

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             638.61      2475.25      2273.27       2500       2296

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             421.00      1560.00      3720.00       1560       3720

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             955.56      3802.02       501.01       3764        496

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             275.76       965.66      4016.16        956       3976

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             853.00      3384.00       840.00       3384        840

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             560.40      2130.69      3223.76       2152       3256

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             500.00      1907.07      2787.88       1888       2760

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             133.00       352.00      5296.00        352       5296

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              53.00         0.00      6244.00          0       6244

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb             267.00       904.00      4840.00        904       4840

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
sdb              52.00         0.00      6240.00          0       6240



-- 
Tomasz Chmielewski
http://wpkg.org

             reply	other threads:[~2009-03-23 15:38 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-03-23 15:38 Tomasz Chmielewski [this message]
2009-03-23 21:23 ` [linux-lvm] why LVM reads when I tell it to write? Larry Dickson

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=49C7ACE1.4070000@wpkg.org \
    --to=mangoo@wpkg.org \
    --cc=linux-lvm@redhat.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.