If you happen to be running a Linux system with Xen support enabled, beware there may be odd behavior with the Linux kernel's power management -- it can easily move in either direction.
Following the publishing of the Xen vs. KVM vs. VirtualBox virtualization benchmarks published last month on Phoronix, a reader had noted something quite interesting within the forums.
"I am currently writing my master's thesis about virtualization and I've been running Xen and KVM tests among other things. I've had similar results with Xen being quite slow - for example, its raw CPU speed in LINPACK test was only 90 % that of HW or KVM (which were very close). I've been using Ubuntu 10.04 and vanilla 3.0.0 kernel in my tests. Just yesterday I did a test with Xen Dom0 kernel 126.96.36.199 and for example the LINPACK result jumped to on par with the hardware. I've measured other quite interesting results also - for example Xen's idle power consumption was like 30 % higher with kernel 3.0.0. It really seems there's something badly wrong in the 3.0.0 kernel with Xen. Next, I'll try running my tests with the 3.2 release candidate kernel to see if things have changed."
In particular, his comment about Xen's power consumption being 30% higher than the normal kernel was found to be quite interesting. Upon running my own set of tests, I too have found that when Xen Dom0 support is present the Linux kernel power management is quite awkward. However, it is not always for the worse.
This testing is looking at the power management between the vanilla Linux kernel and when the Linux kernel is booted with the Xen Hypervisor support enabled, but not looking when running any DomU guests or other changes. It is simply looking at the performance on the host with and without Xen support.
The first Xen 4.1 power usage monitoring tests was done from an Intel Core i5 2520M "Sandy Bridge" Hewlett-Packard notebook.