Ever since the release of Windows 2008 Server I have been hearing the buzz about it being a better Windows XP replacement then Windows Vista. During the first month after its release I attributed this buzz to the normal hype surrounding every release of a new version of Windows and, as is my habit, I ignored it but, the buzz continued.
The buzz continued with a number of articles explaining how to make Windows 2008 Server behave like its desktop counterpart. The first one was an article from Vijayshina Karnure at blogs.msdn.com titled Using Windows Server 2008 as a SUPER workstation OS, which detailed configuring Windows server to behave like Vista. Of course he kept the best parts of Windows 2008 Server enabled. Features like Hyper-V and hardware virtualization. My first reaction was, “Why would anyone want to make Windows 2008 Server look like Vista?” My reason for that reaction was, I feel that Aero might be the source of my performance problems on Vista. To be honest, the blame for that probably falls at the graphics driver’s feet. The second article that caught my eye was Christian Mohn’s in which he explained his experience using Windows 2008 Server as a desktop OS.
From Christian Mohn’s article:
Windows Server 2008 performs better, even with the Aero features enabled, than Vista ever did on the same hardware. To me, this a bit strange, even if a lot of services are still disabled, as the code base is pretty much the same as Vista. For all I know, Vista might perform better now that Service Pack 1 has been released, but I don’t anticipate changing back any time soon. Another factor here is that I had the 32bit version of Vista installed, while I’m now running 64bit Server 2008. How much of a difference that really makes, I don’t know.
The final article that made me take notice was posted at exo-blog.blogspot which detailed the first set of soft numbers I had seen, proving that Windows 2008 server was indeed faster then Vista on the same hardware. The author of that article claimed a “11-17%” increase in performance.
With data now in hand, I decided to try it for myself. I installed Windows Vista SP1 and Windows 2008 Server, configured according to the articles above, on 1.6 Core 2 Duo with 2 GB of RAM and an Intel integrated graphics card. This was by no means a video game system, rather my intent was to see if the difference between the performance of Vista and Windows 2008 Server would hold up on your typical business PC.
The short answer is, it did. There was at least a 10% improvement overall. My test was hardly scientific, consisting of a stop watch but, I did record a real improvement in system boot up and general business application performance. I tested mundane things such as calculation time in a large Excel worksheet and riping a CD using media player.
So, now my question is; “How can Windows 2008 Sever be faster then Vista if they are basically the same code base?”
I took a quick look at the running processes on both machines and nothing stood out. I also did your basic Google search to see if anyone else had come up with an answer. I never found a smoking gun. I guess we will have to wait for the professional testing sites to figure it out.
Either way, my point is, Microsoft should either back port whatever changes were made to the Windows 2008 code base to Vista or, if that isn’t possible, release a desktop version. I think this would go a long way to improving Vista’s public image. Plus a little good press never hurt.
If you have access to Windows 2008 Server, try this for yourself, and lets see if we can figure out how to tweak Vista to match Windows 2008 server.