Port-xen archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: PV drivers in HVM DomU



On 9/28/2011 4:53 AM, Jean-Yves Migeon wrote:
For mono-CPU work, I don't think you will see much difference between
HVM and PV, except for I/O (PV should clearly win). For multi-CPU number
crunching stuff, things will be different (although this depends on the
situation): sometimes, the mono-CPU job is _faster_ that the two-CPUs
one. Lock overhead and contention, maybe.

Up to date results would be interesting. Thanks in advance if you have
some!


Here are the results of my totally unscientific experiment...

Machine: dual Xeon 5148 2.33GHz
PV gets 2048MB RAM and 1 core
HVM gets 2048MB RAM and 4 cores (but emulated I/O)

* DL'd the 5.1 syssrc.tgz and config'd GENERIC

* created test.sh that contained nothing but "make depend && make"

* Ran "time test.sh" in PV, got the following:

>       439.94 real       318.11 user       117.53 sys

* Ran "make clean; make cleandir"

* Ran "time test.sh" in HVM, got the following:

>       638.56 real       385.82 user       226.97 sys

* Ran "make clean; make cleandir"

* Changed "make" to "make -j4" in test.sh

* Ran "time test.sh" in HVM again, got the following:

>      269.94 real        439.90 user       326.49 sys

So, it appears that the emulated I/O is significantly slower as expected, but for this job on my system, that isn't the bottleneck.

-d


Home | Main Index | Thread Index | Old Index