qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Michael R. Hines" <mrhines@linux.vnet.ibm.com>
To: Anthony Liguori <aliguori@us.ibm.com>
Cc: quintela@redhat.com, qemu-devel@nongnu.org, owasserm@redhat.com,
	Bulent Abali <abali@us.ibm.com>,
	Michael R Hines <mrhines@us.ibm.com>,
	Gokul B Kandiraju <gokul@us.ibm.com>,
	pbonzini@redhat.com
Subject: Re: [Qemu-devel] [PATCH v5 00/12] rdma: migration support
Date: Tue, 23 Apr 2013 13:54:35 -0400	[thread overview]
Message-ID: <5176CADB.4000304@linux.vnet.ibm.com> (raw)
In-Reply-To: <87li89dtmc.fsf@codemonkey.ws>

On 04/23/2013 01:50 PM, Anthony Liguori wrote:
> mrhines@linux.vnet.ibm.com writes:
>
>> From: "Michael R. Hines" <mrhines@us.ibm.com>
>>
>> Juan, Please pull.
> I assume this is actually v6, not v5?
>
> I don't see collected Reviewed-bys...
>
> That said, we're pretty close to hard freeze.  I think this should wait
> until 1.6 opens up although I'm open to suggestion if people think this
> is low risk.  I don't like the idea of adding a new protocol this close
> to the end of a cycle.
>
> Regards,
>
> Anthony Liguori

There are no instructions/procedures documented on the qemu.org
website on how to automatically generate "Reviewed-by" signatures.

How do you guys do that?

- Michael

>> Changes since v4:
>>
>> - Re-ran checkpatch.pl
>> - Added new QEMUFileOps function: qemu_get_max_size()
>> - Renamed capability to x-pin-all, disabled by default
>> - Added numbers for x-pin-all to performance section in docs/rdma.txt
>> - Included performance numbers in this cover letter
>> - Converted throughput patch to a MigrationStats statistic in QMP
>> - Better QMP error message delivery
>> - Updated documentation
>> - Moved docs/rdma.txt up to top of patch series
>> - Fixed all v4 changes requested
>> - Finished additional cleanup requests
>> - Updated copyright for migration-rdma.c
>>
>> Wiki: http://wiki.qemu.org/Features/RDMALiveMigration
>> Github: git@github.com:hinesmr/qemu.git
>>
>> Here is a brief summary of total migration time and downtime using RDMA:
>>
>> Using a 40gbps infiniband link performing a worst-case stress test,
>> using an 8GB RAM virtual machine:
>> Using the following command:
>>
>> $ apt-get install stress
>> $ stress --vm-bytes 7500M --vm 1 --vm-keep
>>
>> RESULTS:
>>
>> 1. Migration throughput: 26 gigabits/second.
>> 2. Downtime (stop time) varies between 15 and 100 milliseconds.
>>
>> EFFECTS of memory registration on bulk phase round:
>>
>> For example, in the same 8GB RAM example with all 8GB of memory in
>> active use and the VM itself is completely idle using the same 40 gbps
>> infiniband link:
>>
>> 1. x-rdma-pin-all disabled total time: approximately 7.5 seconds @ 9.5 Gbps
>> 2. x-rdma-pin-all enabled total time: approximately 4 seconds @ 26 Gbps
>>
>> These numbers would of course scale up to whatever size virtual machine
>> you have to migrate using RDMA.
>>
>> Enabling this feature does *not* have any measurable affect on
>> migration *downtime*. This is because, without this feature, all of the
>> memory will have already been registered already in advance during
>> the bulk round and does not need to be re-registered during the successive
>> iteration rounds.
>>
>> Michael R. Hines (12):
>>    rdma: add documentation
>>    rdma: export yield_until_fd_readable()
>>    rdma: export throughput w/ MigrationStats QMP
>>    rdma: introduce qemu_get_max_size()
>>    rdma: introduce qemu_file_mode_is_not_valid()
>>    rdma: export qemu_fflush()
>>    rdma: introduce ram_handle_compressed()
>>    rdma: introduce qemu_ram_foreach_block()
>>    rdma: new QEMUFileOps hooks
>>    rdma: introduce capability x-rdma-pin-all
>>    rdma: core logic
>>    rdma: send pc.ram
>>
>>   Makefile.objs                 |    1 +
>>   arch_init.c                   |   59 +-
>>   configure                     |   29 +
>>   docs/rdma.txt                 |  404 ++++++
>>   exec.c                        |    9 +
>>   hmp.c                         |    2 +
>>   include/block/coroutine.h     |    6 +
>>   include/exec/cpu-common.h     |    5 +
>>   include/migration/migration.h |   24 +
>>   include/migration/qemu-file.h |   44 +
>>   migration-rdma.c              | 2727 +++++++++++++++++++++++++++++++++++++++++
>>   migration.c                   |   22 +-
>>   qapi-schema.json              |   12 +-
>>   qemu-coroutine-io.c           |   23 +
>>   savevm.c                      |  133 +-
>>   15 files changed, 3451 insertions(+), 49 deletions(-)
>>   create mode 100644 docs/rdma.txt
>>   create mode 100644 migration-rdma.c
>>
>> -- 
>> 1.7.10.4

  reply	other threads:[~2013-04-23 17:54 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-23  1:55 [Qemu-devel] [PATCH v5 00/12] rdma: migration support mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 01/12] rdma: add documentation mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 02/12] rdma: export yield_until_fd_readable() mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 03/12] rdma: export throughput w/ MigrationStats QMP mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 04/12] rdma: introduce qemu_get_max_size() mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 05/12] rdma: introduce qemu_file_mode_is_not_valid() mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 06/12] rdma: export qemu_fflush() mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 07/12] rdma: introduce ram_handle_compressed() mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 08/12] rdma: introduce qemu_ram_foreach_block() mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 09/12] rdma: new QEMUFileOps hooks mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 10/12] rdma: introduce capability x-rdma-pin-all mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 11/12] rdma: core logic mrhines
2013-04-23 20:59   ` Paolo Bonzini
2013-04-23 23:53     ` Michael R. Hines
2013-04-24  6:38       ` Paolo Bonzini
2013-04-24 18:19         ` Michael R. Hines
2013-04-23 21:10   ` Paolo Bonzini
2013-04-24  0:01     ` Michael R. Hines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 12/12] rdma: send pc.ram mrhines
2013-04-23 17:50 ` [Qemu-devel] [PATCH v5 00/12] rdma: migration support Anthony Liguori
2013-04-23 17:54   ` Michael R. Hines [this message]
2013-04-23 18:26     ` Anthony Liguori
2013-04-23 18:46       ` Michael R. Hines
2013-04-23 18:55         ` Anthony Liguori
2013-04-23 19:24       ` Eric Blake
2013-04-23 20:15         ` Michael R. Hines
2013-04-23 21:11           ` Eric Blake
2013-04-23 21:44             ` Anthony Liguori
2013-04-23 22:40               ` [Qemu-devel] contribution process [was: [PATCH v5 00/12] rdma: migration support] Eric Blake
2013-04-23 23:50                 ` Michael R. Hines
2013-04-23 23:50               ` [Qemu-devel] [PATCH v5 00/12] rdma: migration support Michael R. Hines
2013-04-23 21:11           ` Paolo Bonzini
2013-04-23 23:49             ` Michael R. Hines
2013-04-23 20:16       ` Michael R. Hines
  -- strict thread matches above, loose matches on Subject: below --
2013-04-21 21:17 mrhines

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=5176CADB.4000304@linux.vnet.ibm.com \
    --to=mrhines@linux.vnet.ibm.com \
    --cc=abali@us.ibm.com \
    --cc=aliguori@us.ibm.com \
    --cc=gokul@us.ibm.com \
    --cc=mrhines@us.ibm.com \
    --cc=owasserm@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@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).