From: Zdenek Kabelac <zkabelac@redhat.com>
To: LVM general discussion and development <linux-lvm@redhat.com>,
Gionatan Danti <g.danti@assyoma.it>,
"Stuart D. Gathman" <stuart@gathman.org>
Cc: "Dalebjörk, Tomas" <tomas.dalebjork@gmail.com>
Subject: Re: [linux-lvm] exposing snapshot block device
Date: Wed, 23 Oct 2019 17:37:22 +0200 [thread overview]
Message-ID: <13d450bf-f6ad-7db8-eec1-88c4eb473287@redhat.com> (raw)
In-Reply-To: <0fde933f-1e8d-60de-cada-4a49a67129c4@assyoma.it>
Dne 23. 10. 19 v 16:37 Gionatan Danti napsal(a):
> On 23/10/19 14:59, Zdenek Kabelac wrote:
>> Dne 23. 10. 19 v 13:08 Gionatan Danti napsal(a):
>>> Talking about thin snapshot, an obvious performance optimization which
>>> seems to not be implemented is to skip reading source data when overwriting
>>> in larger-than-chunksize blocks.
>
> "For example, consider a completely filled 64k chunk thin volume (with
> thinpool having ample free space). Snapshotting it and writing a 4k block on
> origin will obviously cause a read of the original 64k chunk, an in-memory
> change of the 4k block and a write of the entire modified 64k block to a new
> location. But writing, say, a 1 MB block should *not* cause the same read on
> source: after all, the read data will be immediately discarded, overwritten by
> the changed 1 MB block."
>
> I would expect that such large-block *thin* snapshot rewrite behavior would
> not cause a read/modify/write, but it really does.
>
> Is this a low-hanging fruit or there are more fundamental problem avoiding
> read/modify/write in this case?
Hi
If you use 1MiB chunksize for thin-pool and you use 'dd' with proper bs size
and you write 'aligned' on 1MiB boundary (be sure you user directIO, so you
are not a victim of some page cache flushing...) - there should not be any
useless read.
If you still do see such read - and you can easily reproduce this with latest
kernel - report a bug please with your reproducer and results.
Regards
Zdenek
next prev parent reply other threads:[~2019-10-23 15:37 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-22 10:47 [linux-lvm] exposing snapshot block device Dalebjörk, Tomas
2019-10-22 13:57 ` Zdenek Kabelac
2019-10-22 15:29 ` Dalebjörk, Tomas
2019-10-22 15:36 ` Zdenek Kabelac
2019-10-22 16:13 ` Dalebjörk, Tomas
2019-10-23 10:26 ` Zdenek Kabelac
2019-10-23 10:56 ` Tomas Dalebjörk
2019-10-22 16:15 ` Stuart D. Gathman
2019-10-22 17:02 ` Tomas Dalebjörk
2019-10-22 21:38 ` Gionatan Danti
2019-10-22 22:53 ` Stuart D. Gathman
2019-10-23 6:58 ` Gionatan Danti
2019-10-23 10:06 ` Tomas Dalebjörk
2019-10-23 10:12 ` Zdenek Kabelac
2019-10-23 10:46 ` Zdenek Kabelac
2019-10-23 11:08 ` Gionatan Danti
2019-10-23 11:24 ` Tomas Dalebjörk
2019-10-23 11:26 ` Tomas Dalebjörk
2019-10-24 16:01 ` Zdenek Kabelac
2019-10-25 16:31 ` Tomas Dalebjörk
2019-11-04 5:54 ` Tomas Dalebjörk
2019-11-04 10:07 ` Zdenek Kabelac
2019-11-04 14:40 ` Tomas Dalebjörk
2019-11-04 15:04 ` Zdenek Kabelac
2019-11-04 17:28 ` Tomas Dalebjörk
2019-11-05 16:24 ` Zdenek Kabelac
2019-11-05 16:40 ` Mikulas Patocka
2019-11-05 20:56 ` Tomas Dalebjörk
2019-11-06 9:22 ` Zdenek Kabelac
2019-11-07 16:54 ` Mikulas Patocka
2019-11-07 17:29 ` Tomas Dalebjörk
2020-09-04 12:09 ` Tomas Dalebjörk
2020-09-04 12:37 ` Zdenek Kabelac
2020-09-07 13:09 ` Mikulas Patocka
2020-09-07 14:14 ` Dalebjörk, Tomas
2020-09-07 14:17 ` Zdenek Kabelac
2020-09-07 16:34 ` Tomas Dalebjörk
2020-09-07 16:42 ` Zdenek Kabelac
2020-09-07 17:37 ` Tomas Dalebjörk
2020-09-07 17:50 ` Zdenek Kabelac
2020-09-08 12:32 ` Dalebjörk, Tomas
2020-09-07 19:56 ` Tomas Dalebjörk
2020-09-07 20:22 ` Tomas Dalebjörk
2020-09-07 21:02 ` Tomas Dalebjörk
2019-10-23 12:12 ` Ilia Zykov
2019-10-23 12:20 ` Ilia Zykov
2019-10-23 13:05 ` Zdenek Kabelac
2019-10-23 14:40 ` Gionatan Danti
2019-10-23 15:46 ` Ilia Zykov
2019-10-23 12:59 ` Zdenek Kabelac
2019-10-23 14:37 ` Gionatan Danti
2019-10-23 15:37 ` Zdenek Kabelac [this message]
2019-10-23 17:16 ` Gionatan Danti
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=13d450bf-f6ad-7db8-eec1-88c4eb473287@redhat.com \
--to=zkabelac@redhat.com \
--cc=g.danti@assyoma.it \
--cc=linux-lvm@redhat.com \
--cc=stuart@gathman.org \
--cc=tomas.dalebjork@gmail.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 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).