linux-lvm.redhat.com archive mirror
 help / color / mirror / Atom feed
From: Gionatan Danti <g.danti@assyoma.it>
To: LVM general discussion and development <linux-lvm@redhat.com>
Cc: Arvid Picciani <aep@exys.org>
Subject: Re: [linux-lvm] is lvmraid 1 thinpool with integrity safe?
Date: Tue, 11 Apr 2023 11:15:33 +0200	[thread overview]
Message-ID: <03be4a57a1ef162918fc906c76682db5@assyoma.it> (raw)
In-Reply-To: <CACD9L=A9i6QFzXtnQh6e3=gD9u6cy3VJAtCEaf1WsoCU_eE3kQ@mail.gmail.com>

Il 2023-04-09 14:21 Arvid Picciani ha scritto:
> Hi,
> 
> doing some performance tests i noticed that lvmraid + integrity +
> thinpool outperforms zfs z1 by 5x while offering the same features.
> (snapshots, integrity)
> 
> Is this somehow unsafe or how come it is so unpopular?
> 
> lvcreate --type raid1 --mirrors 1 --size '100M' loop --name loopmeta
> --raidintegritymode journal
> lvcreate --type raid1 --mirrors 1 --size '700M' loop --name loopdata
> --raidintegritymode journal
> lvconvert --type thin-pool --poolmetadata loop/loopmeta loop/loop

I benchmarked this setup quite heavily some years ago (when RHEL 8 was 
shiny new), using integritysetup because lvm integration was not 
available.

Be aware that in journal mode all random writes are sequentially 
accumulated in the journal and are later destaged. If using HDDs this 
last part can block your array for extended time. For example, using 
fio, I noticed 50k random writes for 2s followed by 30s where random 
iops was 0 (zero). Bitmap mode (rather than journal) was better 
speed-wise, but be sure to understand the tradeoff.

Other notes regarding ZFS:
- ZRAID1 is the equivalent of RAID5 and has lower performance than 
mirrors;
- recordsize plays an important role, but can affect benchmark results 
in a non-intuitive manner;
- ZFS is heavily dependent on ARC - don't try to outsmart it with 
syntethical tests, rather benchmark it your real workload.

Regards.

-- 
Danti Gionatan
Supporto Tecnico
Assyoma S.r.l. - www.assyoma.it
email: g.danti@assyoma.it - info@assyoma.it
GPG public key ID: FF5F32A8

_______________________________________________
linux-lvm mailing list
linux-lvm@redhat.com
https://listman.redhat.com/mailman/listinfo/linux-lvm
read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/


  reply	other threads:[~2023-04-11  9:23 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-09 12:21 [linux-lvm] is lvmraid 1 thinpool with integrity safe? Arvid Picciani
2023-04-11  9:15 ` Gionatan Danti [this message]
2023-04-11 13:15 ` David Teigland

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=03be4a57a1ef162918fc906c76682db5@assyoma.it \
    --to=g.danti@assyoma.it \
    --cc=aep@exys.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 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).