Virtualization Company Logos Sun.com Intel.com AMD.com HP.com RedHat.com Apple.com SWSoft.com VMware.com Parallels.com Microsoft.com cj tracking image cj tracking image
Does Google Really Have Their Head In The Clouds?
September 19th, 2008 under x86 Virtualization, Open Source, VMworld, Google.com, Virtualization, Desktop Computing, Enterprise Computing, Microsoft

Just returning from VMworld, the idea of cloud computing fresh on my mind, I come across these two articles, “A Web OS? Are You Dense?” By Ted Dziuba and the original article he references “Meet Chrome, Google’s Windows Killer” by Michael Arrington. I think they both missed the point, of what Google is doing.

Cloud computing is moving the computation and storage of data from individual machines or servers, into a dynamic cloud where resources are allocated on a dynamic level instantaneously based on actually needs of the calling systems at that time. Previous computational models, had storage either done locally or centrally, and similar distributions of computational power, as the price of desktop hardware fluctuates over time in connection with Moore’s Law.

Google Branded Pano Box

Both of these authors are short sighted in reviewing Google’s plans. They are looking at currently technology, and the complexity which has evolved over time, to get to where we are now. They didn’t look far enough ahead to see where Google is going. After having the opportunity to review some very slick thin client devices, from Pano Logic, Inc, Wyse Technology and others at VMworld 2008, let me offer you my take.

Here is a description of Pano Logic’s device, direct from their website, read what the device doesn’t have in BOLD and replace the words “Microsoft Windows” with “Google Chrome”:

Pano is purpose-built for server-based desktop virtualization. Moving all software off the desktop to the server, the Pano device has no CPU, no memory, no operating system and no drivers. Using functionality enabled by server virtualization, Pano delivers not only a complete Microsoft Windows experience, including full USB support, but also a new set of high-value features not available with desktop PC and thin client architectures. (source)

What Google is aiming to do, is transition from the current model, where you have desktop applications using local data and storing it locally, to a future model where the computational operations and data storage and backup happens within a cloud. You now sit down a terminal, and log into your Google world, accessing every document you may have ever written, every e-mail you have every read, every website you have ever visited, connected to their cloud from a dumb terminal, such as the Pano Logic box. The advantages over time for cloud computing are huge. It is a very green option, as this becomes more of an issue for enterprise clients, who have tens of thousands of systems running 24/7. Replace these desktop computers with their 350 watt power supplies with a dumb terminal which can have little as a 5 watt power supply and the power savings alone are huge, before even looking at loss and theft, cooling, upgrades, etc over time.

What is the problem with Chrome? Besides the initial issues with the EULA, it is trying to do revolutionary things on non revolutionary hardware. Windows is a bloated mess, which was designed for local execution and local storage of data. It doesn’t like taking a back seat and playing chauffeur to a cloud based computing client. I would like to show the illustration which Ted so nicely crafted, with his own unique style and flair, but I feel the language is too course for this site. What he does illustrate is the layers which data has to pass to go from the hardware up to the application. What he missed in his analysis, is that chrome is a transitional product, as Google devices become available, which will use protocols still under development, such as net2display, their will be no local operating system layers to process. The Google device will be a tv set top box, that simple displays images rendered on Google Farms and sent to the terminal via Ethernet, wifi or wimax technology. The device then returns mouse and keyboard feedback allowing Google to process the input. This may sound far fetched, but it really isn’t that far off. I can use my Apple ipod touch, in combination with RDP software to view and control my desktop computer from 5 states away and get acceptable user experience overall.

A closing quote, from Sergey Brin, co-founder of Google, “As we go forward, I hope we’re going to continue to use technology to make really big differences in how people live and work.” (source)

Resources:

ORGs for Scalable, Robust, Privacy-Friendly Client Cloud Computing

Meet Chrome, Google’s Windows Killer by Michael Arrington

A Web OS? Are You Dense? By Ted Dziuba


No related posts


One Response to “Does Google Really Have Their Head In The Clouds?”

  1. Deep Into the Cloud Computing « Todor Ivanov’s Weblog Says:

    […] last but not list one interesting post discussing Google’s Cloud approach from x86Virtualization […]

Leave a Reply