Re: vCPU scalability for linux VMs

2010-05-06 Thread Alec Istomin
On Wednesday, May 5, 2010 at 01:53:55 -0400, Srivatsa Vaddagiri wrote: On Wed, May 05, 2010 at 12:31:11PM -0700, Alec Istomin wrote: On Wednesday, May 5, 2010 at 13:27:39 -0400, Srivatsa Vaddagiri wrote: My preliminary results show that single vCPU Linux VMs perform up to 10 times better

Re: vCPU scalability for linux VMs

2010-05-05 Thread Alec Istomin
On Wednesday, May 5, 2010 at 04:43:32 -0400, Avi Kivity wrote: So you have a total of 32 vcpus on 8 cores? This is known to be problematic. You may see some improvement by enabling hyperthreading. exactly, 32 vCPUs on 8 core hardware that doesn't support hyperthreading (Clovertown E5335)

Re: vCPU scalability for linux VMs

2010-05-05 Thread Alec Istomin
On Wednesday, May 5, 2010 at 13:27:39 -0400, Srivatsa Vaddagiri wrote: My preliminary results show that single vCPU Linux VMs perform up to 10 times better than 4vCPU Linux VMs (consolidated performance of 8 VMs on 8 core pre-Nehalem server). I suspect that I'm missing something major and

vCPU scalability for linux VMs

2010-05-04 Thread Alec Istomin
Gentlemen, Reaching out with a non-development question, sorry if it's not appropriate here. I'm looking for a way to improve Linux SMP VMs performance under KVM. My preliminary results show that single vCPU Linux VMs perform up to 10 times better than 4vCPU Linux VMs (consolidated

Re: how to tweak kernel to get the best out of kvm?

2010-04-23 Thread Alec Istomin
Hi there! I'm trying to understand kvm performance bottleneck and I was hoping to get sharing statistics form ksm. Was not able to locate it so far, the below mentioned /sys entry is missing, while ksm module is loaded. I'm using latest from rhel 5 x64: 2.6.18-194.el5, kmod-kvm-83-164.el5 and I'm