Well, I had this thought myself. Does VMWare have anything to worry about now that Microsoft's Windows Server 2008 is going to have the Hyper-V virtualization bundled with the OS?

History has shown (Netscape) that as Windows bundles software with an OS, that some companies can dissappear.

Here is an article I was reading that covered this topic..

Virtualization Review | Feature: Analysis: Taking Aim at VMware

VMWare doesn't seem too concerned.. At the printing of the article they claim almost 75% of virtualized Windows Servers are running their architecture.

It's a good read.

__________________
Admin
Virtualization

Not too keen on that link admin - I clicked on it and my AV went crazy, claimed lots of trojans were trying to load from that page - after 5 mins it just crashed my IE

MIght want to check it - could just be my PC, but you never know

Hmmm, Strange... The link works fine for me. Does anyone else have a problem with it?

__________________
Admin
Virtualization

The really interesting part of this discussion centers around what Microsoft is going to do with Cloud Computing - that is, computing delivered from remote data centers to users on a pay-for-what-you-use basis. So far, that type of service has been delivered solely on VMWare technology. When you add VMWare and Windows licensing costs, users of Cloud Computing end up paying over double for equivalent server power compared to linux-based services such as those from Amazon, Layered Tech, ENKI, etc. or similar services such as Joyent or Terremark.

The promise of Windows-in-the-cloud is to reduce costs companies pay for Internet-facing Windows servers, but can Microsoft deliver? Right now, Microsoft is trying to control access to "paravirtualized I/O drivers" that are necessary to run Windows under Xen. Since Xen is open source, this type of solution would reduce costs to end users significantly. It looks like Microsoft is positioning itself to be a Windows Cloud Computing provider and trying to lock competitors out of the game.

There are a couple things that could easily derail Microsoft. First, the paravirtualized drivers are being written by third parties and are now available without going through Microsoft (or Citrix, which owns Xensoft, which is the commercial Xen provider and provides the underlying technology for Microsoft's virtualization play.) This will open up the market for Microsoft-based services in the cloud. Secondly, the Microsoft platform is simply more expensive than linux, even if the costs of virtualization are excluded: the licenses for windows and SQL server are pricey compared to the free Linux and Postgres or MySQL. The "elasticity" (scalability) of cloud computing is a mismatch for Windows' one-license-per-server model, requiring cloud providers to pre-allocate licenses to unused servers while they wait for customers to use them, which pushes costs up as well. Also, much of the application software for Linux is for-pay versus open source for Linux. So, I feel the Cloud Computing market will cause Microsoft's market share to go down over time.

However, if Microsoft can own the entire stack of hardware and software and provide it as a service, they can bypass the licensing issues and offer Cloud Computing on Windows as a service. This is where I think they're headed. And it may work to eliminate some competition from VMWare and Cloud hosting services based on it. However, by the time they get the pricing right, they may have missed the market window for Windows in the Cloud if a suffiently large ecosystem of users develops for linux-based cloud services.

___________
Eric Novikoff
ENKI