From: Orit Wasserman <owasserm@redhat.com>
To: Peter Lieven <pl@kamp.de>
Cc: Stefan Hajnoczi <stefanha@gmail.com>,
Paolo Bonzini <pbonzini@redhat.com>,
qemu-devel@nongnu.org, quintela@redhat.com
Subject: Re: [Qemu-devel] [PATCHv4 9/9] migration: use XBZRLE only after bulk stage
Date: Mon, 25 Mar 2013 12:16:10 +0200 [thread overview]
Message-ID: <515023EA.6060103@redhat.com> (raw)
In-Reply-To: <1363956370-23681-10-git-send-email-pl@kamp.de>
On 03/22/2013 02:46 PM, Peter Lieven wrote:
> at the beginning of migration all pages are marked dirty and
> in the first round a bulk migration of all pages is performed.
>
> currently all these pages are copied to the page cache regardless
> of whether they are frequently updated or not. this doesn't make sense
> since most of these pages are never transferred again.
>
> this patch changes the XBZRLE transfer to only be used after
> the bulk stage has been completed. that means a page is added
> to the page cache the second time it is transferred and XBZRLE
> can benefit from the third time of transfer.
>
> since the page cache is likely smaller than the number of pages
> it's also likely that in the second round the page is missing in the
> cache due to collisions in the bulk phase.
>
> on the other hand a lot of unnecessary mallocs, memdups and frees
> are saved.
>
> the following results have been taken earlier while executing
> the test program from docs/xbzrle.txt. (+) with the patch and (-)
> without. (thanks to Eric Blake for reformatting and comments)
>
> + total time: 22185 milliseconds
> - total time: 22410 milliseconds
>
> Shaved 0.3 seconds, better than 1%!
>
> + downtime: 29 milliseconds
> - downtime: 21 milliseconds
>
> Not sure why downtime seemed worse, but probably not the end of the world.
>
> + transferred ram: 706034 kbytes
> - transferred ram: 721318 kbytes
>
> Fewer bytes sent - good.
>
> + remaining ram: 0 kbytes
> - remaining ram: 0 kbytes
> + total ram: 1057216 kbytes
> - total ram: 1057216 kbytes
> + duplicate: 108556 pages
> - duplicate: 105553 pages
> + normal: 175146 pages
> - normal: 179589 pages
> + normal bytes: 700584 kbytes
> - normal bytes: 718356 kbytes
>
> Fewer normal bytes...
>
> + cache size: 67108864 bytes
> - cache size: 67108864 bytes
> + xbzrle transferred: 3127 kbytes
> - xbzrle transferred: 630 kbytes
>
> ...and more compressed pages sent - good.
>
> + xbzrle pages: 117811 pages
> - xbzrle pages: 21527 pages
> + xbzrle cache miss: 18750
> - xbzrle cache miss: 179589
>
> And very good improvement on the cache miss rate.
>
> + xbzrle overflow : 0
> - xbzrle overflow : 0
>
> Signed-off-by: Peter Lieven <pl@kamp.de>
> Reviewed-by: Eric Blake <eblake@redhat.com>
> ---
> arch_init.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch_init.c b/arch_init.c
> index b2b932a..86f7e28 100644
> --- a/arch_init.c
> +++ b/arch_init.c
> @@ -464,7 +464,7 @@ static int ram_save_block(QEMUFile *f, bool last_stage)
> acct_info.skipped_pages++;
> bytes_sent = 0;
> }
> - } else if (migrate_use_xbzrle()) {
> + } else if (!ram_bulk_stage && migrate_use_xbzrle()) {
> current_addr = block->offset + offset;
> bytes_sent = save_xbzrle_page(f, p, current_addr, block,
> offset, cont, last_stage);
>
Reviewed-by: Orit Wasserman <owasserm@redhat.com>
next prev parent reply other threads:[~2013-03-25 10:15 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-22 12:46 [Qemu-devel] [PATCHv4 0/9] buffer_is_zero / migration optimizations Peter Lieven
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 1/9] move vector definitions to qemu-common.h Peter Lieven
2013-03-25 8:35 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 2/9] cutils: add a function to find non-zero content in a buffer Peter Lieven
2013-03-22 19:37 ` Eric Blake
2013-03-22 20:03 ` Peter Lieven
2013-03-22 20:22 ` [Qemu-devel] indentation hints [was: [PATCHv4 2/9] cutils: add a function to find non-zero content in a buffer] Eric Blake
2013-03-23 11:18 ` Peter Maydell
2013-03-25 8:53 ` [Qemu-devel] [PATCHv4 2/9] cutils: add a function to find non-zero content in a buffer Orit Wasserman
2013-03-25 8:56 ` Peter Lieven
2013-03-25 9:26 ` Orit Wasserman
2013-03-25 9:42 ` Paolo Bonzini
2013-03-25 10:03 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 3/9] buffer_is_zero: use vector optimizations if possible Peter Lieven
2013-03-25 8:53 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 4/9] bitops: use vector algorithm to optimize find_next_bit() Peter Lieven
2013-03-25 9:04 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 5/9] migration: search for zero instead of dup pages Peter Lieven
2013-03-22 19:49 ` Eric Blake
2013-03-22 20:02 ` Peter Lieven
2013-03-25 9:30 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 6/9] migration: add an indicator for bulk state of ram migration Peter Lieven
2013-03-25 9:32 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 7/9] migration: do not sent zero pages in bulk stage Peter Lieven
2013-03-22 20:13 ` Eric Blake
2013-03-25 9:44 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 8/9] migration: do not search dirty " Peter Lieven
2013-03-25 10:05 ` Orit Wasserman
2013-03-22 12:46 ` [Qemu-devel] [PATCHv4 9/9] migration: use XBZRLE only after " Peter Lieven
2013-03-25 10:16 ` Orit Wasserman [this message]
2013-03-22 17:25 ` [Qemu-devel] [PATCHv4 0/9] buffer_is_zero / migration optimizations Paolo Bonzini
2013-03-22 19:20 ` Peter Lieven
2013-03-22 21:24 ` Paolo Bonzini
2013-03-23 7:34 ` Peter Lieven
2013-03-25 10:17 ` Peter Lieven
2013-03-25 10:53 ` Paolo Bonzini
2013-03-25 11:26 ` Peter Lieven
2013-03-25 13:02 ` Paolo Bonzini
2013-03-25 13:23 ` Peter Lieven
2013-03-25 13:32 ` Peter Lieven
2013-03-25 14:34 ` Paolo Bonzini
2013-03-25 21:37 ` Peter Lieven
2013-03-26 8:14 ` Peter Lieven
2013-03-26 9:20 ` Paolo Bonzini
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=515023EA.6060103@redhat.com \
--to=owasserm@redhat.com \
--cc=pbonzini@redhat.com \
--cc=pl@kamp.de \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=stefanha@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).