-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Andrew Lentvorski wrote:
> Defend that statement with some stats, please.

The paper I am most familiar with is:

http://www.cl.cam.ac.uk/netos/papers/2003-xensosp.pdf

and then:

http://www.clarkson.edu/class/cs644/xen/files/repeatedxen-usenix04.pdf

which successfully reproduces the claims made in the first paper.

> I've seen claims both ways but very little hard data.  VMWare can do 
> some very clever stuff that avoids kernel trips/page faults altogether. 

They can't possibly avoid these things for disk/network IO, right? All
of that will have to involve some emulation by the hypervisor. In the
benchmarks in the paper above you will note that vmware does far worse
when faced with a workload that involves lots of IO. This is due to the
emulation of privileged instructions required. I do not see any way in
which emulating those can possibly be as fast. Xen gets around this by
getting help from the guest OS. They are likely to be much more closely
matched in a system with hardware emulation.

- --
Tracy R Reed                  Read my blog at http://ultraviolet.org
Key fingerprint = D4A8 4860 535C ABF8 BA97  25A6 F4F2 1829 9615 02AD
Non-GPG signed mail gets read only if I can find it among the spam.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFGX7ZO9PIYKZYVAq0RAgfEAKCKAEzd78Ojz0BCs+DlnYTAkvvDtQCdF1Bs
E4qL9AnA3rBsnjiiDlSW3CQ=
=KnEm
-----END PGP SIGNATURE-----


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to