Avi Kivity wrote:
On 04/12/2010 05:04 AM, Zhang, Xiantao wrote:PLE has the ability to eliminate the issue at some extent, and pv solution should be helpful also. But for windows guests running on machines without PLE, we still needs to enhance host side to resolve the issue.
Is this on a machine that does pause-loop exits? The current handingWhat was the performance hit? What was your I/O setup (imageThe issue only happens when vcpu number is over-committed(e.g.
format, using aio?)
vcpu/pcpu>2) and physical cpus are saturated. For example, when run
webbench in windows OS in this case, its performance drops by 80%.
In our experiment, we are using image file through virtio, and I
think aio should be used by default also.
of PLE is very suboptimal. With proper directed yield we should be
much better there.
Without PLE, we need paravirtualized spinlocks, no way around it.
Spin loops need to be addressed first, they are known to killEven in overcommit case, if vcpu threads of one qemu are not scheduled or pulled to the same logical processor, the performance drop is tolerant like Xen's case today. But for KVM, it has to suffer from additional performance loss, since host's scheduler actively pulls these vcpu threads together.
performance in overcommit situations.