Google recently posted a blog highlighting its plans to release the Chrome OS for both x86 and ARM CPUs in 2010. Google is making this move to provide mobile devices like netbooks and smart phones and with a fast and lightweight OS that gets people onto the web in a few seconds. The assumption is that the user experience takes place “in the cloud” – not on the client computing device and the OS is the vehicle to get you there.

A large number of articles and blogs have popped up on the web as the result of this blog. Since everyone loves a fight – everyone is picking a pairing. Some focus on a Google versus Microsoft battle – others on the x86 versus ARM. And of course this provides new fire power for the on-going open source versus proprietary software discussions since the Chrome OS is described as a basic Linux kernel with a new windowing system and the Chrome browser.

I believe the real issue revolves around where applications are executed. The bottom line – somewhere you need  CPU and GPU processing power  to execute the demanding applications  and deliver the high-end visual experience we all have come to love.

In the Chrome OS world the entire computational action takes place in server farms with the client device providing a window into this world- it does not execute applications or store data.  ”A netbook running Chrome OS isn’t likely to be able to do much when it is not connected to the Internet”, said Al Gillen, an analyst at IDC in a recent article by the Wall Street Journal “Google Targets Microsoft’s Turf.” The Chrome OS is banking on a fast and pervasive adoption of the Cloud Computing model – the web as the delivery vehicle for applications, data, and services.

This is in contrast to today’s most popular desktop, laptops, and mobile devices – most of which can surf the web and also execute applications. What does this mean? These devices have functions even when off-line – like playing games, watching a movie, listening to music, editing photos, or even writing a blog. This is the world where Microsoft Windows runs the majority of devices. In the government and commercial market, which accounts for a majority of PC sales, 70% of applications require Microsoft Windows, Gartner analyst Michael Silver estimates in another Wall Street Journal article “Chrome May Not Shine for Google.” That is why Microsoft is refining the client virtualization model in the upcoming Windows 7 OS to ensure that you – the end user ─ can continue to execute your favorite applications.

So who wins the battle of where applications execute? My crystal ball is cloudy today so I don’t have any predictions – but I believe we are in for an interesting time of innovation. I believe we’re moving toward server farms that can deliver impressive processing power of CPU and GPUs, client devices that can drive richer visual experience, and more and more demanding on-line application environments (take a look at my “Thunder Clouds” post to read about Facebook and some of their requirements).

The HPC world has been using the concept of compute farms to drive applications for years – check out the blog by my colleague Andy Parma wrote called “Is the Cloud a Cluster or is Cluster a Cloud” for interesting viewpoint on how the world is blurring. 

And here is a fun fact – the first OS used for real work is generally thought to be GM-NAA I/O, produced in 1956 by the General Motors research division for its IBM mainframe ─ in an era where OS were mainly developed by the end user!

Margaret Lewis, director, AMD commercial software and solutions

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

@margaretjlewis