qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@gmail.com>
To: Dietmar Maurer <dietmar@proxmox.com>
Cc: Kevin Wolf <kwolf@redhat.com>, Fam Zheng <famz@redhat.com>,
	qemu-devel <qemu-devel@nongnu.org>,
	"imain@redhat.com" <imain@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Wenchao Xia <xiawenc@linux.vnet.ibm.com>
Subject: Re: [Qemu-devel] [PATCH v3 0/8] block: drive-backup live backup command
Date: Thu, 23 May 2013 10:04:17 +0200	[thread overview]
Message-ID: <20130523080417.GC2921@stefanha-thinkpad.redhat.com> (raw)
In-Reply-To: <24E144B8C0207547AD09C467A8259F755859A87E@lisa.maurer-it.com>

On Wed, May 22, 2013 at 03:34:18PM +0000, Dietmar Maurer wrote:
> > That sounds like more work than a persistent dirty bitmap.  The advantage is that
> > while dirty bitmaps are consumed by a single user, the Merkle tree can be used
> > to sync up any number of replicas.
> 
> I also consider it safer, because you make sure the data exists (using hash keys like SHA1).
> 
> I am unsure how you can check if a dirty bitmap contains errors, or is out of date?
> 
> Also, you can compare arbitrary Merkle trees, whereas a dirty bitmap is always related to a single image.
> (consider the user remove the latest backup from the backup target). 

One disadvantage of Merkle trees is that the client becomes stateful -
the client needs to store its own Merkle tree and this requires fancier
client-side code.

It is also more expensive to update hashes than a dirty bitmap.  Not
because you need to hash data but because a small write (e.g. 1 sector)
requires that you read the surrounding sectors to recompute a hash for
the cluster.  Therefore you can expect worse guest I/O performance than
with a dirty bitmap.

I still think it's a cool idea.  Making it work well will require a lot
more effort than a dirty bitmap.

Stefan

  reply	other threads:[~2013-05-23  8:04 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-15 14:34 [Qemu-devel] [PATCH v3 0/8] block: drive-backup live backup command Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 1/8] block: add bdrv_add_before_write_cb() Stefan Hajnoczi
2013-05-15 14:42   ` Paolo Bonzini
2013-05-16  2:42     ` Wenchao Xia
2013-05-16  8:11     ` Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 2/8] block: add basic backup support to block driver Stefan Hajnoczi
2013-05-16  3:27   ` Wenchao Xia
2013-05-16  7:30     ` Stefan Hajnoczi
2013-05-20 11:30   ` Paolo Bonzini
2013-05-21 13:34     ` Stefan Hajnoczi
2013-05-21 15:11       ` Paolo Bonzini
2013-05-21 15:25         ` Dietmar Maurer
2013-05-21 15:35           ` Paolo Bonzini
2013-05-21 15:54             ` Dietmar Maurer
2013-05-21 16:03               ` Paolo Bonzini
2013-05-21 16:15                 ` Dietmar Maurer
2013-05-21 16:26         ` Dietmar Maurer
2013-05-21 16:46           ` Paolo Bonzini
2013-05-22 13:37             ` Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 3/8] block: add drive-backup QMP command Stefan Hajnoczi
2013-05-15 19:04   ` Eric Blake
2013-05-16  7:31     ` Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 4/8] qemu-iotests: add 055 drive-backup test case Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 5/8] blockdev: rename BlkTransactionStates to singular Stefan Hajnoczi
2013-05-15 19:09   ` Eric Blake
2013-05-16  2:28   ` Wenchao Xia
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 6/8] blockdev: add DriveBackup transaction Stefan Hajnoczi
2013-05-15 19:13   ` Eric Blake
2013-05-16  7:36     ` Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 7/8] blockdev: add Abort transaction Stefan Hajnoczi
2013-05-15 19:01   ` Eric Blake
2013-05-16  2:26     ` Wenchao Xia
2013-05-16  7:37     ` Stefan Hajnoczi
2013-05-15 14:34 ` [Qemu-devel] [PATCH v3 8/8] qemu-iotests: test 'drive-backup' transaction in 055 Stefan Hajnoczi
2013-05-16  6:16 ` [Qemu-devel] [PATCH v3 0/8] block: drive-backup live backup command Wenchao Xia
2013-05-16  7:47   ` Stefan Hajnoczi
2013-05-17  6:58     ` Wenchao Xia
2013-05-17  9:14       ` Stefan Hajnoczi
2013-05-21  3:25         ` Wenchao Xia
2013-05-21  7:34           ` Stefan Hajnoczi
2013-05-17 10:17     ` Paolo Bonzini
2013-05-20  6:24       ` Stefan Hajnoczi
2013-05-20  7:23         ` Paolo Bonzini
2013-05-21  7:31           ` Stefan Hajnoczi
2013-05-21  8:30             ` Paolo Bonzini
2013-05-21 10:34               ` Stefan Hajnoczi
2013-05-21 10:36                 ` Paolo Bonzini
2013-05-21 10:58                   ` Dietmar Maurer
2013-05-22 13:43                     ` Stefan Hajnoczi
2013-05-22 15:10                       ` Dietmar Maurer
2013-05-22 15:34                       ` Dietmar Maurer
2013-05-23  8:04                         ` Stefan Hajnoczi [this message]
2013-05-23  8:11                           ` Dietmar Maurer
2013-05-24  8:38                             ` Stefan Hajnoczi
2013-05-24  9:53                               ` Dietmar Maurer
  -- strict thread matches above, loose matches on Subject: below --
2013-09-02 12:57 Benoît Canet
2013-09-03  7:54 ` Stefan Hajnoczi

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=20130523080417.GC2921@stefanha-thinkpad.redhat.com \
    --to=stefanha@gmail.com \
    --cc=dietmar@proxmox.com \
    --cc=famz@redhat.com \
    --cc=imain@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=xiawenc@linux.vnet.ibm.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).