All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Lieven <pl@kamp.de>
To: qemu-devel@nongnu.org
Cc: kwolf@redhat.com, pbonzini@redhat.com, Peter Lieven <pl@kamp.de>
Subject: [Qemu-devel] [PATCHv3 9/9] migration: use XBZRLE only after bulk stage
Date: Thu, 21 Mar 2013 16:57:37 +0100	[thread overview]
Message-ID: <1363881457-14814-10-git-send-email-pl@kamp.de> (raw)
In-Reply-To: <1363881457-14814-1-git-send-email-pl@kamp.de>

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
if there 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
its 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>
---
 arch_init.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch_init.c b/arch_init.c
index 4718d39..4e89783 100644
--- a/arch_init.c
+++ b/arch_init.c
@@ -450,7 +450,7 @@ static int ram_save_block(QEMUFile *f, bool last_stage)
                     qemu_put_byte(f, 0);
                 }
                 bytes_sent++;
-            } 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);
-- 
1.7.9.5

  parent reply	other threads:[~2013-03-21 15:58 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-21 15:57 [Qemu-devel] [PATCHv3 0/9] buffer_is_zero / migration optimizations Peter Lieven
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 1/9] move vector definitions to qemu-common.h Peter Lieven
2013-03-21 17:29   ` Eric Blake
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 2/9] cutils: add a function to find non-zero content in a buffer Peter Lieven
2013-03-21 18:12   ` Eric Blake
2013-03-21 19:11     ` Peter Lieven
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 3/9] buffer_is_zero: use vector optimizations if possible Peter Lieven
2013-03-21 18:16   ` Eric Blake
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 4/9] bitops: use vector algorithm to optimize find_next_bit() Peter Lieven
2013-03-21 19:18   ` Eric Blake
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 5/9] migration: search for zero instead of dup pages Peter Lieven
2013-03-21 19:24   ` Eric Blake
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 6/9] migration: add an indicator for bulk state of ram migration Peter Lieven
2013-03-21 19:27   ` Eric Blake
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 7/9] migration: do not sent zero pages in bulk stage Peter Lieven
2013-03-21 19:26   ` Eric Blake
2013-03-21 19:44     ` Peter Lieven
2013-03-21 15:57 ` [Qemu-devel] [PATCHv3 8/9] migration: do not search dirty " Peter Lieven
2013-03-21 19:27   ` Eric Blake
2013-03-21 19:57     ` Peter Lieven
2013-03-21 15:57 ` Peter Lieven [this message]
2013-03-21 19:31   ` [Qemu-devel] [PATCHv3 9/9] migration: use XBZRLE only after " Eric Blake

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=1363881457-14814-10-git-send-email-pl@kamp.de \
    --to=pl@kamp.de \
    --cc=kwolf@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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.