archive mirror
 help / color / mirror / Atom feed
From: Demi Marie Obenour <>
Subject: Re: [linux-lvm] Need inputs: Performance issues with LVM snapshots
Date: Mon, 7 Mar 2022 09:49:11 -0500	[thread overview]
Message-ID: <> (raw)
In-Reply-To: <>

[-- Attachment # Type: text/plain, Size: 1871 bytes --]

On 3/7/22 09:44, Gionatan Danti wrote:
> Il 2022-03-07 12:09 Gaikwad, Hemant ha scritto:
>> Hi,
>> We have been looking at LVM as an option for long term backup using
>> the LVM snapshots. After looking at the various forums and also
>> looking at a few LVM defects, realized that LVM could be a very good
>> option for short term backup, but might result in performance issues
>> if the snapshots are retained for a long time. Also read we should
>> restrict the number of snapshots. We are thinking of keeping it to 3,
>> but do you think that could also be a performance bottleneck. A few
>> forum posts also suggest memory issues with using LVM snapshots. Can
>> you please help with some data on that too. Thanks in advance for
>> making our decision easier. Thanks
>> Regards,
> Classical, non-thin LVM snapshots are only meant to be short-lived (just 
> enough to take a backup), and the performance penalty you talk about 
> does apply.
> Thin LVM snapshots, on the other side, command a much lower performance 
> penalty and can be long-lived (ie: think about a rolling snapshot 
> system).
> So if you need multiple, long-lived snapshots, I strongly suggest you to 
> check lvmthin.
> Regards.

Also worth noting that there is a minimum time to perform an LVM
operation of any type, a bit over 0.2 seconds on my machine.  If you
need to create snapshots exceedingly quickly, then LVM itself will be a
bottleneck.  Virtually all applications will not run into this problem,
however.  The only ones I can think of that will are container or VM
managers that need to spin up a lot of containers or VMs, as Qubes
OS does.  The time I mentioned is the time needed to run the entire LVM
command; the time I/O is suspended for is far, *far* shorter.

Demi Marie Obenour (she/her/hers)
Invisible Things Lab

[-- Attachment # OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 4963 bytes --]

[-- Attachment #1.2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 202 bytes --]

linux-lvm mailing list
read the LVM HOW-TO at

      reply	other threads:[~2022-03-07 14:49 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-07 11:09 [linux-lvm] Need inputs: Performance issues with LVM snapshots Gaikwad, Hemant
2022-03-07 14:44 ` Gionatan Danti
2022-03-07 14:49   ` Demi Marie Obenour [this message]

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:

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \ \ \ \

* 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).