Hi, Avi
>
> Yes, some cloud vendors already knew that different PLE values has big
> performance impact on their applications. They want one interface for them to
> set. And I think the big cloud vendors should have administrators that have
> experience on PLE tuning. :-)
>
For current stage, d
> -Original Message-
> From: Avi Kivity [mailto:a...@redhat.com]
> Sent: Wednesday, October 17, 2012 6:12 PM
> To: Hu, Xuekun
> Cc: k...@vger.kernel.org; qemu-devel@nongnu.org; Zhang, Xiantao
> Subject: Re: [Patch]KVM: enabling per domain PLE
>
> On 10/17/2012 10:02 AM, Hu, Xuekun wrote:
On 10/17/2012 10:02 AM, Hu, Xuekun wrote:
>>
>> The problem with this is that it requires an administrator to understand the
>> workload, not only of the guest, but also of other guests on the machine.
>> With low overcommit, a high PLE window reduces unneeded exits, but with
>> high overcommit we
>
> The problem with this is that it requires an administrator to understand the
> workload, not only of the guest, but also of other guests on the machine.
> With low overcommit, a high PLE window reduces unneeded exits, but with
> high overcommit we need those exits to reduce spinning.
>
> In a
On 10/16/2012 08:53 AM, Hu, Xuekun wrote:
> Setting the same PLE parameter arbitrarily for different
> workloads is not a good solution.
True.
> The solution enables
> per domain PLE which gives user ability to set PLE parameter
> for different domain for better performance.
The problem with th
Setting the same PLE parameter arbitrarily for different
workloads is not a good solution. The solution enables
per domain PLE which gives user ability to set PLE parameter
for different domain for better performance.
Signed-off-by: Xuekun Hu
---
arch/x86/include/asm/kvm_host.h |6 ++
ar
