From: Ori Mamluk <omamluk@zerto.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: "Kevin Wolf" <kwolf@redhat.com>,
"תומר בן אור" <tomer@zertodata.com>, "עודד קדם" <oded@zerto.com>,
dlaor@redhat.com, qemu-devel@nongnu.org,
"Yair Kuszpet" <yairk@zerto.com>,
"Paolo Bonzini" <pbonzini@redhat.com>
Subject: [Qemu-devel] [RFC] Replication agent requirements (was [RFC PATCH] replication agent module)
Date: Wed, 08 Feb 2012 15:00:19 +0200 [thread overview]
Message-ID: <4F3271E3.50704@zerto.com> (raw)
In-Reply-To: <CAJSP0QVkmrmzwWygc66unytA9BHA5gENr0tFa1Urp3WxMeDb1w@mail.gmail.com>
Hi,
Following previous mails from Kevin and Dor, I'd like to specify the
high level requirements of a replication agent as I see them.
1. Report each write to a protected volume to the rephub, at an IO
transaction granularity
* The reporting is not synchronous, i.e. the write completion is
not delayed until the rephub received it.
* The IOs have to be the raw guest IOs - i.e. not converted to any
sparse format or another filter that alters the size/offset
2. Report failures to report an IO (e.g. socket disconnect or send
timeout) or failed IOs (bad status from storage) to rephub
* It is enough to disconnect the socket - that can be considered a
'failure report'
3. Enable rephub to read arbitrary regions in the protected volume
* Assume that rephub can identify IOs which were dropped by the
replication system, and needs to re-read the data of these IOs.
We'd like to treat the following requirement as a second stage - not to
implement it in the first version:
4. Synchronously report IO writes meta data (offset, size) to an
external API
* Synchronous meaning that it is reported (blocking) before the IO
is processed by storage.
* The goal is to maintain a dirty bitmap outside of the Qemu process
* The tracking needs to be more persistent than the Qemu process. A
good example for that is to expose an additional process API
(yet another NBD??) that will be hold the bitmap by either the
host RAM or by writing persistently to storage.
The emphasis to report single IO transactions is because high end
replication (near synchronous) requires access to every IO shortly after
it is written to the storage.
Thanks,
Ori
next prev parent reply other threads:[~2012-02-08 13:00 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-07 10:29 [Qemu-devel] [RFC PATCH] replication agent module Ori Mamluk
2012-02-07 12:12 ` Anthony Liguori
2012-02-07 12:25 ` Dor Laor
2012-02-07 12:30 ` Ori Mamluk
2012-02-07 12:40 ` Anthony Liguori
2012-02-07 14:06 ` Ori Mamluk
2012-02-07 14:40 ` Paolo Bonzini
2012-02-07 14:48 ` Ori Mamluk
2012-02-07 15:47 ` Paolo Bonzini
2012-02-08 6:10 ` Ori Mamluk
2012-02-08 8:49 ` Dor Laor
2012-02-08 11:59 ` Stefan Hajnoczi
2012-02-08 8:55 ` Kevin Wolf
2012-02-08 9:47 ` Ori Mamluk
2012-02-08 10:04 ` Kevin Wolf
2012-02-08 13:28 ` [Qemu-devel] [RFC] Replication agent design (was [RFC PATCH] replication agent module) Ori Mamluk
2012-02-08 14:59 ` Stefan Hajnoczi
2012-02-08 14:59 ` Stefan Hajnoczi
2012-02-19 13:40 ` Ori Mamluk
2012-02-20 14:32 ` Paolo Bonzini
2012-02-21 9:03 ` [Qemu-devel] BlockDriverState stack and BlockListeners (was: [RFC] Replication agent design) Kevin Wolf
2012-02-21 9:15 ` [Qemu-devel] BlockDriverState stack and BlockListeners Paolo Bonzini
2012-02-21 9:49 ` Kevin Wolf
2012-02-21 10:09 ` Paolo Bonzini
2012-02-21 10:51 ` Kevin Wolf
2012-02-21 11:36 ` Paolo Bonzini
2012-02-21 12:22 ` Stefan Hajnoczi
2012-02-21 12:57 ` Paolo Bonzini
2012-02-21 15:49 ` Markus Armbruster
2012-02-21 13:10 ` Kevin Wolf
2012-02-21 13:21 ` Paolo Bonzini
2012-02-21 15:56 ` Markus Armbruster
2012-02-21 16:04 ` Kevin Wolf
2012-02-21 16:19 ` Markus Armbruster
2012-02-21 16:39 ` Kevin Wolf
2012-02-21 17:16 ` Stefan Hajnoczi
2012-02-21 10:20 ` Ori Mamluk
2012-02-29 8:38 ` Ori Mamluk
2012-03-03 11:46 ` Stefan Hajnoczi
2012-03-04 5:14 ` Ori Mamluk
2012-03-04 8:56 ` Paolo Bonzini
2012-03-05 12:04 ` Stefan Hajnoczi
2012-02-08 11:02 ` [Qemu-devel] [RFC PATCH] replication agent module Stefan Hajnoczi
2012-02-08 13:00 ` Ori Mamluk [this message]
2012-02-08 13:30 ` [Qemu-devel] [RFC] Replication agent requirements (was [RFC PATCH] replication agent module) Anthony Liguori
2012-02-08 12:03 ` [Qemu-devel] [RFC PATCH] replication agent module Stefan Hajnoczi
2012-02-08 12:46 ` Paolo Bonzini
2012-02-08 14:39 ` Stefan Hajnoczi
2012-02-08 14:55 ` Paolo Bonzini
2012-02-08 15:07 ` Stefan Hajnoczi
2012-02-07 14:53 ` Kevin Wolf
2012-02-07 15:00 ` Anthony Liguori
2012-02-07 13:34 ` Kevin Wolf
2012-02-07 13:50 ` Stefan Hajnoczi
2012-02-07 13:58 ` Paolo Bonzini
2012-02-07 14:05 ` Paolo Bonzini
2012-02-08 12:17 ` Orit Wasserman
2012-02-07 14:18 ` Ori Mamluk
2012-02-07 14:59 ` Anthony Liguori
2012-02-07 15:20 ` Stefan Hajnoczi
2012-02-07 16:25 ` Anthony Liguori
2012-02-21 16:01 ` Markus Armbruster
2012-02-21 17:31 ` Stefan Hajnoczi
2012-02-07 14:45 ` Ori Mamluk
2012-02-08 12:29 ` Orit Wasserman
2012-02-08 11:45 ` Luiz Capitulino
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=4F3271E3.50704@zerto.com \
--to=omamluk@zerto.com \
--cc=dlaor@redhat.com \
--cc=kwolf@redhat.com \
--cc=oded@zerto.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@gmail.com \
--cc=tomer@zertodata.com \
--cc=yairk@zerto.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.