From: Gionatan Danti <g.danti@assyoma.it>
To: Mark Mielke <mark.mielke@gmail.com>
Cc: LVM general discussion and development <linux-lvm@redhat.com>
Subject: Re: [linux-lvm] Snapshot behavior on classic LVM vs ThinLVM
Date: Sat, 08 Apr 2017 13:56:50 +0200 [thread overview]
Message-ID: <3772958ec700a4f7e1b152a97e605e0f@assyoma.it> (raw)
In-Reply-To: <CALm7yL1Qm4xanZ=tbLgC7B7KZ=dzLZtsTJXTd0rsEONxg0nqng@mail.gmail.com>
Il 08-04-2017 00:24 Mark Mielke ha scritto:
>
> We use lvmthin in many areas... from Docker's dm-thinp driver, to XFS
> file systems for PostgreSQL or other data that need multiple
> snapshots, including point-in-time backup of certain snapshots. Then,
> multiple sizes. I don't know that we have 8 TB anywhere right this
> second, but we are using it in a variety of ranges from 20 GB to 4 TB.
>
Very interesting, this is the exact information I hoped to get. Thank
you for reporting.
>
> When you say "nightly", my experience is that processes are writing
> data all of the time. If the backup takes 30 minutes to complete, then
> this is 30 minutes of writes that get accumulated, and subsequent
> performance overhead of these writes.
>
> But, we usually keep multiple hourly snapshots and multiply daily
> snapshots, because we want the option to recover to different points
> in time. With the classic LVM snapshot capability, I believe this is
> essentially non-functional. While it can work with "1 short lived
> snapshot", I don't think it works at all well for "3 hourly + 3 daily
> snapshots". Remember that each write to an area will require that
> area to be replicated multiple times under classic LVM snapshots,
> before the original write can be completed. Every additional snapshot
> is an additional cost.
Right. For such a setup, classic LVM snapshot overhead would be
enormous, grinding all to an halt.
>
>> I more concerned about lenghtly snapshot activation due to a big,
>> linear CoW table that must be read completely...
>
> I suspect this is a pre-optimization concern, in that you are
> concerned, and you are theorizing about impact, but perhaps you
> haven't measured it yourself, and if you did, you would find there was
> no reason to be concerned. :-)
For classic (non-thinly provided) LVM snapshot, relatively big metadata
size was a know problem. Many talks happened on this list for this very
topic. Basically, when the snapshot metadata size increased above a
certain point (measured in some GB), snapshot activation failed due to
timeout on LVM commands. This, in turn, was due that legacy snapshot
behavior was not really tuned for long-lived, multi-gigabyte snapshots,
rather for create-backup-remove behavior.
>
> If you absolutely need a contiguous sequence of blocks for your
> drives, because your I/O patterns benefit from this, or because your
> hardware has poor seek performance (such as, perhaps a tape drive? :-)
> ), then classic LVM snapshots would retain this ordering for the live
> copy, and the snapshot could be as short lived as possible to minimize
> overhead to only that time period.
>
> But, in practice - I think the LVM authors of the thinpool solution
> selected a default block size that would exhibit good behaviour on
> most common storage solutions. You can adjust it, but in most cases I
> think I don't bother, and just use the default. There is also the
> behaviour of the systems in general to take into account in that even
> if you had a purely contiguous sequence of blocks, your file system
> probably allocates files all over the drive anyways. With XFS, I
> believe they do this for concurrency, in that two different kernel
> threads can allocate new files without blocking each other, because
> they schedule the writes to two different areas of the disk, with
> separate inode tables.
>
> So, I don't believe the contiguous sequence of blocks is normally a
> real thing. Perhaps a security camera that is recording a 1+ TB video
> stream might allocate contiguous, but basically nothing else does
> this.
True.
>
> To me, LVM thin volumes is the right answer to this problem. It's not
> particularly new or novel either. Most "Enterprise" level storage
> systems have had this capability for many years. At work, we use
> NetApp and they take this to another level with their WAFL =
> Write-Anywhere-File-Layout. For our private cloud solution based upon
> NetApp AFF 8080EX today, we have disk shelves filled with flash
> drives, and NetApp is writing everything "forwards", which extends the
> life of the flash drives, and allows us to keep many snapshots of the
> data. But, it doesn't have to be flash to take advantage of this. We
> also have large NetApp FAS 8080EX or 8060 with all spindles, including
> 3.5" SATA disks. I was very happy to see this type of technology make
> it back into LVM. I think this breathed new life into LVM, and made it
> a practical solution for many new use cases beyond being just a more
> flexible partition manager.
>
> --
>
> Mark Mielke <mark.mielke@gmail.com>
Yeah, CoW-enabled filesystem are really cool ;) Too bad BTRFS has very
low performance when used as VM backing store...
Thank you very much Mark, I really appreciate the information you
provided.
--
Danti Gionatan
Supporto Tecnico
Assyoma S.r.l. - www.assyoma.it
email: g.danti@assyoma.it - info@assyoma.it
GPG public key ID: FF5F32A8
next prev parent reply other threads:[~2017-04-08 11:56 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-06 14:31 [linux-lvm] Snapshot behavior on classic LVM vs ThinLVM Gionatan Danti
2017-04-07 8:19 ` Mark Mielke
2017-04-07 9:12 ` Gionatan Danti
2017-04-07 13:50 ` L A Walsh
2017-04-07 16:33 ` Gionatan Danti
2017-04-13 12:59 ` Stuart Gathman
2017-04-13 13:52 ` Xen
2017-04-13 14:33 ` Zdenek Kabelac
2017-04-13 14:47 ` Xen
2017-04-13 15:29 ` Stuart Gathman
2017-04-13 15:43 ` Xen
2017-04-13 17:26 ` Stuart D. Gathman
2017-04-13 17:32 ` Stuart D. Gathman
2017-04-14 15:17 ` Xen
2017-04-14 7:27 ` Gionatan Danti
2017-04-14 7:23 ` Gionatan Danti
2017-04-14 15:23 ` Xen
2017-04-14 15:53 ` Gionatan Danti
2017-04-14 16:08 ` Stuart Gathman
2017-04-14 17:36 ` Xen
2017-04-14 18:59 ` Gionatan Danti
2017-04-14 19:20 ` Xen
2017-04-15 8:27 ` Xen
2017-04-15 23:35 ` Xen
2017-04-17 12:33 ` Xen
2017-04-15 21:22 ` Xen
2017-04-15 21:49 ` Xen
2017-04-15 21:48 ` Xen
2017-04-18 10:17 ` Zdenek Kabelac
2017-04-18 13:23 ` Gionatan Danti
2017-04-18 14:32 ` Stuart D. Gathman
2017-04-19 7:22 ` Xen
2017-04-07 22:24 ` Mark Mielke
2017-04-08 11:56 ` Gionatan Danti [this message]
2017-04-07 18:21 ` Tomas Dalebjörk
2017-04-13 10:20 ` Gionatan Danti
2017-04-13 12:41 ` Xen
2017-04-14 7:20 ` Gionatan Danti
2017-04-14 8:24 ` Zdenek Kabelac
2017-04-14 9:07 ` Gionatan Danti
2017-04-14 9:37 ` Zdenek Kabelac
2017-04-14 9:55 ` Gionatan Danti
2017-04-22 7:14 ` Gionatan Danti
2017-04-22 16:32 ` Xen
2017-04-22 20:58 ` Gionatan Danti
2017-04-22 21:17 ` Zdenek Kabelac
2017-04-23 5:29 ` Xen
2017-04-23 9:26 ` Zdenek Kabelac
2017-04-24 21:02 ` Xen
2017-04-24 21:59 ` Zdenek Kabelac
2017-04-26 7:26 ` Gionatan Danti
2017-04-26 7:42 ` Zdenek Kabelac
2017-04-26 8:10 ` Gionatan Danti
2017-04-26 11:23 ` Zdenek Kabelac
2017-04-26 13:37 ` Gionatan Danti
2017-04-26 14:33 ` Zdenek Kabelac
2017-04-26 16:37 ` Gionatan Danti
2017-04-26 18:32 ` Stuart Gathman
2017-04-26 19:24 ` Stuart Gathman
2017-05-02 11:00 ` Gionatan Danti
2017-05-12 13:02 ` Gionatan Danti
2017-05-12 13:42 ` Joe Thornber
2017-05-14 20:39 ` Gionatan Danti
2017-05-15 12:50 ` Zdenek Kabelac
2017-05-15 14:48 ` Gionatan Danti
2017-05-15 15:33 ` Zdenek Kabelac
2017-05-16 7:53 ` Gionatan Danti
2017-05-16 10:54 ` Zdenek Kabelac
2017-05-16 13:38 ` Gionatan Danti
2018-02-27 18:39 ` Xen
2018-02-28 9:26 ` Zdenek Kabelac
2018-02-28 19:07 ` Gionatan Danti
2018-02-28 21:43 ` Zdenek Kabelac
2018-03-01 7:14 ` Gionatan Danti
2018-03-01 8:31 ` Zdenek Kabelac
2018-03-01 9:43 ` Gianluca Cecchi
2018-03-01 11:10 ` Zdenek Kabelac
2018-03-01 9:52 ` Gionatan Danti
2018-03-01 11:23 ` Zdenek Kabelac
2018-03-01 12:48 ` Gionatan Danti
2018-03-01 16:00 ` Zdenek Kabelac
2018-03-01 16:26 ` Gionatan Danti
2018-03-03 18:32 ` Xen
2018-03-04 20:34 ` Zdenek Kabelac
2018-03-03 18:17 ` Xen
2018-03-04 20:53 ` Zdenek Kabelac
2018-03-05 9:42 ` Gionatan Danti
2018-03-05 10:18 ` Zdenek Kabelac
2018-03-05 14:27 ` Gionatan Danti
2018-03-03 17:52 ` Xen
2018-03-04 23:27 ` Zdenek Kabelac
2017-04-22 21:22 ` Zdenek Kabelac
2017-04-24 13:49 ` Gionatan Danti
2017-04-24 14:48 ` Zdenek Kabelac
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=3772958ec700a4f7e1b152a97e605e0f@assyoma.it \
--to=g.danti@assyoma.it \
--cc=linux-lvm@redhat.com \
--cc=mark.mielke@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).