thehypervisor.com reports VMware Workstation faster than Hyper-V
August 2nd, 2008 | 
Author: Hyper-V benchmark surprise is a post at thehypervisor.com that shows some interesting performance benchmarks when comparing VMware Workstation with Microsoft Hyper-V.
“In head to head benchmark tests of Microsoft Hyper-V compared to VMware Workstation, The Hypervisor test results indicate virtual machine (VM) performance using Hyper-V is below that of VMware Workstation.
We tested Hyper-V running on a desktop PC fitted with an Intel Core Quad Q6600 CPU running at 2.4GHz. We compared the results with the same PC running Microsoft Vista SP1 x86 and VMware Workstation 6. Our results show VMware Workstation 6 is a faster platform for both Windows and Linux applications, although Hyper-V is almost as fast as VMware Workstation for running Windows Vista.”
Thehypervisor.com concludes the following from their testing:
“Our analysis of these results is that Hyper-V is a reasonable platform for hosting modern Windows software such as Windows Server 2008 and Windows Vista. However, care is needed when using Hyper-V to host older versions of Windows or Linux software, as the performance of virtual machines could be a lot less than is possible using alternative virtualisation products.”
In my opinion, these results support Microsoft’s normal strategic thinking of doing everything possible to get their customers to adopt their new OS and phasing out support for the previous versions as quickly as possible. I’m sure this post will be “gasoline on the fire” for those who are already up in arms about the attempts to phase out Windows XP support. I’ll stir the flames – check out the full post for the specific XP installation benchmark.
I discovered this post while reading the DABCC.com post Microsoft’s Enterprise Virtualization Solution, Hyper-V, Slower than VMware’s Desktop Version (Workstation) where Douglas Brown sums up my similar reaction to these findings:
“This is really amazing as Hyper-V is Microsoft’s “enterprise” virtualization solution and VMware Workstation is just a “lowly” desktop virtualization that is not suppose to be faster than a “bare metal hypervisor”. Go Figure…”
I’m now curious how VMware Server and ESXi benchmarks would compare to Hyper-V in these same tests.
Related Posts
Posted in hyper-v, vmware, workstation | 
Tags: esx, hyper-v, microsoft, virtualization, vmware, vmware workstation
that is a pretty pointless post on their part. What is the point in timing the installation routine; specifically when at that point all the devices are emulated. Hyper-V performs best with the Integration Services installed in each VM. As does VMware with the tools installed.
Keith,
I understand your point, and I had a similar debate with myself about reporting these benchmarks on VM /ETC. In the end I decided that thehypervisor.com post was significant because Hyper-V is being marketed as a free, enterprise production ready alternative to ESX. The tests were also fair in that both products were configured consistently – ie no guest tools, performing same tasks, and running on the same hardware. If the reality is that a hosted virtualization product designed for desktop testing and development out performs Hyper-V in any capacity it’s worth talking about.
Hopefully, we will soon see benchmark tests where Domain Controllers, Messaging Servers, Database Servers, Terminal Servers, etc. are compared head to head on the various virtualization platforms with the proper, optimized configuration of each product (installations of guest tools, etc)
Wake me up when they try a meaningful benchmark. I don’t install XP for a living.
Steve,
Agreed. However, if VDI takes off your OS installation responsibilities may change. Then again, Windows 7 will probably be the OS of choice soon enough.
Who knows, MS will probably develop a “Windows VM” OS tailored, and licensed, for VDI. If I can think of it then I’m sure somebody in Redmond is way ahead of me!
I don’t think the tests were fair.
VMware has been built around the “emulate everything” way of thinking. This is not a bad thing, and it has served them well in the past. They’ve been able to offer a very broad range of supported guest OSes, and they’ve done that very well so far.
Hyper-V is built around synthetic devices to provide improved performance. This also means that their emulation has not been optimized heavily for performance.
Of course this benchmark still serves a purpose: If you plan on running OS on Hyper-V that do not have ICs, expect worse performance.