From: Wei Wang <wei.w.wang@intel.com>
To: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
virtualization@lists.linux-foundation.org, kvm@vger.kernel.org,
linux-mm@kvack.org, mst@redhat.com, mhocko@kernel.org,
akpm@linux-foundation.org
Cc: torvalds@linux-foundation.org, pbonzini@redhat.com,
liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
quan.xu0@gmail.com, nilal@redhat.com, riel@redhat.com,
peterx@redhat.com
Subject: Re: [PATCH v34 0/4] Virtio-balloon: support free page reporting
Date: Sat, 30 Jun 2018 12:31:19 +0800 [thread overview]
Message-ID: <5B370797.1090700@intel.com> (raw)
In-Reply-To: <1529928312-30500-1-git-send-email-wei.w.wang@intel.com>
On 06/25/2018 08:05 PM, Wei Wang wrote:
> This patch series is separated from the previous "Virtio-balloon
> Enhancement" series. The new feature, VIRTIO_BALLOON_F_FREE_PAGE_HINT,
> implemented by this series enables the virtio-balloon driver to report
> hints of guest free pages to the host. It can be used to accelerate live
> migration of VMs. Here is an introduction of this usage:
>
> Live migration needs to transfer the VM's memory from the source machine
> to the destination round by round. For the 1st round, all the VM's memory
> is transferred. From the 2nd round, only the pieces of memory that were
> written by the guest (after the 1st round) are transferred. One method
> that is popularly used by the hypervisor to track which part of memory is
> written is to write-protect all the guest memory.
>
> This feature enables the optimization by skipping the transfer of guest
> free pages during VM live migration. It is not concerned that the memory
> pages are used after they are given to the hypervisor as a hint of the
> free pages, because they will be tracked by the hypervisor and transferred
> in the subsequent round if they are used and written.
>
> * Tests
> - Test Environment
> Host: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
> Guest: 8G RAM, 4 vCPU
> Migration setup: migrate_set_speed 100G, migrate_set_downtime 2 second
>
> - Test Results
> - Idle Guest Live Migration Time (results are averaged over 10 runs):
> - Optimization v.s. Legacy = 284ms vs 1757ms --> ~84% reduction
According to Michael's comments, add one more set of data here:
Enabling page poison with value=0, and enable KSM.
The legacy live migration time is 1806ms (averaged across 10 runs),
compared to the case with this optimization feature in use (i.e. 284ms),
there is still around ~84% reduction.
Best,
Wei
WARNING: multiple messages have this Message-ID (diff)
From: Wei Wang <wei.w.wang@intel.com>
To: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
virtualization@lists.linux-foundation.org, kvm@vger.kernel.org,
linux-mm@kvack.org, mst@redhat.com, mhocko@kernel.org,
akpm@linux-foundation.org
Cc: yang.zhang.wz@gmail.com, riel@redhat.com, quan.xu0@gmail.com,
liliang.opensource@gmail.com, pbonzini@redhat.com,
nilal@redhat.com, torvalds@linux-foundation.org
Subject: Re: [PATCH v34 0/4] Virtio-balloon: support free page reporting
Date: Sat, 30 Jun 2018 12:31:19 +0800 [thread overview]
Message-ID: <5B370797.1090700@intel.com> (raw)
In-Reply-To: <1529928312-30500-1-git-send-email-wei.w.wang@intel.com>
On 06/25/2018 08:05 PM, Wei Wang wrote:
> This patch series is separated from the previous "Virtio-balloon
> Enhancement" series. The new feature, VIRTIO_BALLOON_F_FREE_PAGE_HINT,
> implemented by this series enables the virtio-balloon driver to report
> hints of guest free pages to the host. It can be used to accelerate live
> migration of VMs. Here is an introduction of this usage:
>
> Live migration needs to transfer the VM's memory from the source machine
> to the destination round by round. For the 1st round, all the VM's memory
> is transferred. From the 2nd round, only the pieces of memory that were
> written by the guest (after the 1st round) are transferred. One method
> that is popularly used by the hypervisor to track which part of memory is
> written is to write-protect all the guest memory.
>
> This feature enables the optimization by skipping the transfer of guest
> free pages during VM live migration. It is not concerned that the memory
> pages are used after they are given to the hypervisor as a hint of the
> free pages, because they will be tracked by the hypervisor and transferred
> in the subsequent round if they are used and written.
>
> * Tests
> - Test Environment
> Host: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
> Guest: 8G RAM, 4 vCPU
> Migration setup: migrate_set_speed 100G, migrate_set_downtime 2 second
>
> - Test Results
> - Idle Guest Live Migration Time (results are averaged over 10 runs):
> - Optimization v.s. Legacy = 284ms vs 1757ms --> ~84% reduction
According to Michael's comments, add one more set of data here:
Enabling page poison with value=0, and enable KSM.
The legacy live migration time is 1806ms (averaged across 10 runs),
compared to the case with this optimization feature in use (i.e. 284ms),
there is still around ~84% reduction.
Best,
Wei
WARNING: multiple messages have this Message-ID (diff)
From: Wei Wang <wei.w.wang@intel.com>
To: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
virtualization@lists.linux-foundation.org, kvm@vger.kernel.org,
linux-mm@kvack.org, mst@redhat.com, mhocko@kernel.org,
akpm@linux-foundation.org
Cc: torvalds@linux-foundation.org, pbonzini@redhat.com,
liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
quan.xu0@gmail.com, nilal@redhat.com, riel@redhat.com,
peterx@redhat.com
Subject: [virtio-dev] Re: [PATCH v34 0/4] Virtio-balloon: support free page reporting
Date: Sat, 30 Jun 2018 12:31:19 +0800 [thread overview]
Message-ID: <5B370797.1090700@intel.com> (raw)
In-Reply-To: <1529928312-30500-1-git-send-email-wei.w.wang@intel.com>
On 06/25/2018 08:05 PM, Wei Wang wrote:
> This patch series is separated from the previous "Virtio-balloon
> Enhancement" series. The new feature, VIRTIO_BALLOON_F_FREE_PAGE_HINT,
> implemented by this series enables the virtio-balloon driver to report
> hints of guest free pages to the host. It can be used to accelerate live
> migration of VMs. Here is an introduction of this usage:
>
> Live migration needs to transfer the VM's memory from the source machine
> to the destination round by round. For the 1st round, all the VM's memory
> is transferred. From the 2nd round, only the pieces of memory that were
> written by the guest (after the 1st round) are transferred. One method
> that is popularly used by the hypervisor to track which part of memory is
> written is to write-protect all the guest memory.
>
> This feature enables the optimization by skipping the transfer of guest
> free pages during VM live migration. It is not concerned that the memory
> pages are used after they are given to the hypervisor as a hint of the
> free pages, because they will be tracked by the hypervisor and transferred
> in the subsequent round if they are used and written.
>
> * Tests
> - Test Environment
> Host: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
> Guest: 8G RAM, 4 vCPU
> Migration setup: migrate_set_speed 100G, migrate_set_downtime 2 second
>
> - Test Results
> - Idle Guest Live Migration Time (results are averaged over 10 runs):
> - Optimization v.s. Legacy = 284ms vs 1757ms --> ~84% reduction
According to Michael's comments, add one more set of data here:
Enabling page poison with value=0, and enable KSM.
The legacy live migration time is 1806ms (averaged across 10 runs),
compared to the case with this optimization feature in use (i.e. 284ms),
there is still around ~84% reduction.
Best,
Wei
---------------------------------------------------------------------
To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org
next prev parent reply other threads:[~2018-06-30 4:27 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-25 12:05 [PATCH v34 0/4] Virtio-balloon: support free page reporting Wei Wang
2018-06-25 12:05 ` [virtio-dev] " Wei Wang
2018-06-25 12:05 ` Wei Wang
2018-06-25 12:05 ` [PATCH v34 1/4] mm: support to get hints of free page blocks Wei Wang
2018-06-25 12:05 ` [virtio-dev] " Wei Wang
2018-06-25 12:05 ` Wei Wang
2018-06-25 12:05 ` [PATCH v34 2/4] virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT Wei Wang
2018-06-25 12:05 ` [virtio-dev] " Wei Wang
2018-06-26 1:37 ` Michael S. Tsirkin
2018-06-26 1:37 ` [virtio-dev] " Michael S. Tsirkin
2018-06-26 1:37 ` Michael S. Tsirkin
2018-06-26 3:46 ` Wei Wang
2018-06-26 3:46 ` [virtio-dev] " Wei Wang
2018-06-26 3:46 ` Wei Wang
2018-06-26 3:56 ` Michael S. Tsirkin
2018-06-26 3:56 ` [virtio-dev] " Michael S. Tsirkin
2018-06-26 3:56 ` Michael S. Tsirkin
2018-06-26 12:27 ` Wei Wang
2018-06-26 12:27 ` [virtio-dev] " Wei Wang
2018-06-26 12:27 ` Wei Wang
2018-06-26 13:34 ` Michael S. Tsirkin
2018-06-26 13:34 ` [virtio-dev] " Michael S. Tsirkin
2018-06-27 1:24 ` Wei Wang
2018-06-27 1:24 ` [virtio-dev] " Wei Wang
2018-06-27 1:24 ` Wei Wang
2018-06-27 2:41 ` Michael S. Tsirkin
2018-06-27 2:41 ` [virtio-dev] " Michael S. Tsirkin
2018-06-27 2:41 ` Michael S. Tsirkin
2018-06-27 3:00 ` Wei Wang
2018-06-27 3:00 ` [virtio-dev] " Wei Wang
2018-06-27 3:00 ` Wei Wang
2018-06-27 3:58 ` Michael S. Tsirkin
2018-06-27 3:58 ` [virtio-dev] " Michael S. Tsirkin
2018-06-27 5:27 ` Wei Wang
2018-06-27 5:27 ` [virtio-dev] " Wei Wang
2018-06-27 5:27 ` Wei Wang
2018-06-27 16:53 ` [virtio-dev] " Michael S. Tsirkin
2018-06-27 16:53 ` Michael S. Tsirkin
2018-06-27 16:53 ` Michael S. Tsirkin
2018-06-27 3:58 ` Michael S. Tsirkin
2018-06-26 13:34 ` Michael S. Tsirkin
2018-06-25 12:05 ` Wei Wang
2018-06-25 12:05 ` [PATCH v34 3/4] mm/page_poison: expose page_poisoning_enabled to kernel modules Wei Wang
2018-06-25 12:05 ` Wei Wang
2018-06-25 12:05 ` [virtio-dev] " Wei Wang
2018-06-25 12:05 ` [PATCH v34 4/4] virtio-balloon: VIRTIO_BALLOON_F_PAGE_POISON Wei Wang
2018-06-25 12:05 ` [virtio-dev] " Wei Wang
2018-06-25 12:05 ` Wei Wang
2018-06-27 11:06 ` [PATCH v34 0/4] Virtio-balloon: support free page reporting David Hildenbrand
2018-06-27 11:06 ` [virtio-dev] " David Hildenbrand
2018-06-27 11:06 ` David Hildenbrand
2018-06-29 3:51 ` Wei Wang
2018-06-29 3:51 ` [virtio-dev] " Wei Wang
2018-06-29 3:51 ` Wei Wang
2018-06-29 7:46 ` David Hildenbrand
2018-06-29 7:46 ` [virtio-dev] " David Hildenbrand
2018-06-29 7:46 ` David Hildenbrand
2018-06-29 11:31 ` Wei Wang
2018-06-29 11:31 ` [virtio-dev] " Wei Wang
2018-06-29 11:31 ` Wei Wang
2018-06-29 11:53 ` David Hildenbrand
2018-06-29 11:53 ` [virtio-dev] " David Hildenbrand
2018-06-29 11:53 ` David Hildenbrand
2018-06-29 15:55 ` Wang, Wei W
2018-06-29 15:55 ` [virtio-dev] " Wang, Wei W
2018-06-29 15:55 ` Wang, Wei W
2018-06-29 15:55 ` Wang, Wei W
2018-06-29 16:03 ` David Hildenbrand
2018-06-29 16:03 ` [virtio-dev] " David Hildenbrand
2018-06-29 16:03 ` David Hildenbrand
2018-06-29 16:03 ` David Hildenbrand
2018-06-29 14:45 ` Michael S. Tsirkin
2018-06-29 14:45 ` Michael S. Tsirkin
2018-06-29 14:45 ` [virtio-dev] " Michael S. Tsirkin
2018-06-29 15:28 ` David Hildenbrand
2018-06-29 15:28 ` [virtio-dev] " David Hildenbrand
2018-06-29 15:28 ` David Hildenbrand
2018-06-29 15:52 ` Wang, Wei W
2018-06-29 15:52 ` [virtio-dev] " Wang, Wei W
2018-06-29 15:52 ` Wang, Wei W
2018-06-29 15:52 ` Wang, Wei W
2018-06-29 16:32 ` Michael S. Tsirkin
2018-06-29 16:32 ` [virtio-dev] " Michael S. Tsirkin
2018-06-29 16:32 ` Michael S. Tsirkin
2018-06-29 16:32 ` Michael S. Tsirkin
2018-06-30 4:31 ` Wei Wang [this message]
2018-06-30 4:31 ` [virtio-dev] " Wei Wang
2018-06-30 4:31 ` Wei Wang
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=5B370797.1090700@intel.com \
--to=wei.w.wang@intel.com \
--cc=akpm@linux-foundation.org \
--cc=kvm@vger.kernel.org \
--cc=liliang.opensource@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=mst@redhat.com \
--cc=nilal@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=quan.xu0@gmail.com \
--cc=riel@redhat.com \
--cc=torvalds@linux-foundation.org \
--cc=virtio-dev@lists.oasis-open.org \
--cc=virtualization@lists.linux-foundation.org \
--cc=yang.zhang.wz@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 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.