Ubuntu 11.10: Xen vs. KVM vs. VirtualBox

[phoronix]

Phoronix: Ubuntu 11.10: Xen vs. KVM vs. VirtualBox

While last week I showed how Ubuntu's performance has evolved as a KVM guest from Ubuntu 8.04 through Ubuntu 11.10, in today's article is a Linux virtualization showdown between VirtualBox, Xen, and KVM while using Ubuntu 11.10 on the Linux 3.0 kernel.

http://www.phoronix.com/vr.php?view=16625

[disi]

Really good tests, I run VirtualBox for other OS at the moment (mostly as mentioned for 2d/3d support).

The next Linux virtualization benchmarks on Phoronix are seeing where the VirtualBox 3D performance is at for Linux guests. There are also AMD Bulldozer virtualization benchmarks around the corner.

Yes, please

[birdie]

Michael, but where are VMWare Workstation 8 results?

It's a very big omission for such a good research.

[darkbasic]

What's the point to test the ancients Xen HVM paths? PV would have been MUCH more interesting...

[bug77]

The advantage of VirtualBox over KVM, however, is its current support for 2D/3D acceleration within guest virtual machines that is then redirected to the host's graphics card.

In a server environment, I doubt it's much of an advantage. It is nice to have though.

[Scullder]

What's the point to test the ancients Xen HVM paths? PV would have been MUCH more interesting...

Yup, I use Xen, I've never used HVM for a linux guest and I don't get the point of it.
You can also use LVM logical volumes on the host as disks for your guests, and I think it will change a lot of results

[Michael]

Michael, but where are VMWare Workstation 8 results?

It's a very big omission for such a good research.

I think I mentioned it in the article... License agreement doesn't allow me to publish benchmarks...

[kobblestown]

Linux containers in their LXC flavor are a standard feature of current Linux kernels. It would be interesting to see how they stack up. Granted, it's not full virtualization solution but it's worth testing out to see if the benefits in performance outweigh the drawbacks (e.g. the same kernel for all guests).

[DaemonFC]

I think I mentioned it in the article... License agreement doesn't allow me to publish benchmarks...

They get to hide all theiur performance problems under a "thou shalt not benchmark our crappy products or we shall surely sueth thou" clause.

Microsoft and Apple EULAs have the same. Apple EULAs forbid developing nuclear weapons with their software. BSD license lets you benchmark and make nuclear weapons to drop on Australia, but unfortunately BSD licensed code ends up buried under MS and Apple EULAs a lot. (Such as in AMD porting the open source Linux drivers to Windows. Even though this has indirect benefits to open source operating systems, the users of Windows don't get the software under free(dom) or open source terms, and any contributors to AMD's drivers can have their work hijacked and put into Windows even if their goal was only to improve free(dom) and open source software)

So make sure you don't benchmark Internet Explorer DirectX acceleration in Windows running under Parallels in Mac OS X under VMWare while making nuclear weapons to drop on Australia.

At least not in the US, where courts are ruling lately that draconian and arbitrary EULA provisions are enforceable against the end user after all even in violation of fair use, comparison, and the previously upheld first sale doctrine. (See also; Psystar)

No benchmarking clauses should be held under an antitrust microscope since they don't allow the user to decide for themselves which product is superior.

[Michael]

No benchmarking clauses should be held under an antitrust microscope since they don't allow the user to decide for themselves which product is superior.

Their clauses generally just mandate no publishing of benchmarks, not that they can't be carried out for your own personal research.... Granted VMware doesn't seem too bad, they even use the Phoronix Test Suite and OpenBenchmarking.org when publishing some of their own results.