Hi Ted,
>On Wed, Sep 12, 2018 at 12:39:26AM +0000, Wanpeng Li wrote:
>> The kvm hypervisor is at the core of cloud computing, Virtual
>> Machine based approaches to workload consolidation, as seen in IaaS
>> cloud as well as data center platforms, have long had to contend
>> with performance degradation caused by synchronization primitives
>> inside the guest environments. These primitives can be affected by
>> virtual CPU preemptions by the host scheduler that can introduce
>> delays that are orders of magnitude longer than those primitives
>> were designed for. In this presentation, Wanpeng Li will introduce
>> current progress to mitigate TLB Shootdown Preemption. In addition,
>> Wanpeng Li will also introduce Exitless IPIs to scale large VMs.
>
>It looks like you are proposing an English language version of this
>talk from LinuxCon China:
>
>https://lc32018.sched.com/event/ER9W/torwards-a-more-scalable-kvm-hypervisor-wanpeng-li-tencent-cloud
>
>
>LinuxCon + ContainerCon + CloudOpen China 2018: Torwards a more Scalable KVM Hypervisor...
>lc32018.sched.com
>View more about this event at LinuxCon + ContainerCon + CloudOpen China 2018
>
>
>Is that correct?  I wasn't able to find slides from the LinuxCon China
>presentations.  Are they available?
>

Yeah, in attachment. In addition, I hope to replace the contents w/ my TLB Shootdown Preemption 
Mitigation(Paravirtual TLB Shootdown) which can benefit overcommit in the Clouds and Exit-less IPIs
(https://lkml.org/lkml/2018/7/23/108) which can scale large VMs. These features have already been 
merged in upstream this year.

Regards,
Wanpeng Li