Uncommon Wisdom

Dec 8 2010   12:28AM GMT

Google’s Chrome OS: Framework for cloud client device and platform

Tom Nolle Tom Nolle Profile: Tom Nolle

Google let the industry have its first look at its Chrome OS, which it sees as the framework for a “cloud client” device, as well as a platform that combines Google’s desktop position with one in the smartphone space (Android) and a service-side position (Google’s cloud services) to create a new and complete (yes, and completely Google) solution for future computing and communication. I don’t think Google has grandiose visions of owning the computing/networking world, but I do think that it’s thinking through the process of ecosystemic computing more seriously and effectively than most. The commercial launch of Google’s Chrome OS may be delayed, but some of the impacts may be visible even before then.

If you look at Chrome OS and Chrome (the browser on which the OS is based) , what you see is a reflection of the fundamental truth that cloud computing is not ceding computing to the cloud, but rebalancing computing activity between the cloud and the client. In effective cloud computing, the process-intense tasks of information editing and display should be pushed outward to the client to reduce the impact of these tasks on central resources and to insure that the network connection to the client doesn’t become congested with a lot of unnecessary display-oriented babble.

A good example comes from a Google demonstration of the WebGL 3D rendering framework. If you want to show sharks swimming, you can either send pixel-by-pixel information on the successive positions, compressed information on the same thing, or objects that can be locally rendered. The latter will be better than any of the former choices from a cloud performance perspective.

I think it’s clear that’s what Google is thinking. What’s not so clear is whether it can actually achieve its goals of creating cloud dominance, and if it does, whether it can monetize its success there. Some of Google’s recent ventures, like Nexus S and Editions, seem to be a bit less than half-baked in terms of maturity of the business plan. Conceptually, Chrome OS has been around for a long time, and so have Google’s cloud aspirations. That the two were related is no secret, but it’s how they might relate in a business sense that’s hard to see. Does Google think it can sell ads to enterprises that display during their work-day applications? Seems doubtful. Does Google then think that Chrome OS and its cloud approach is a consumer solution to computing? If so, why make such a fuss about things like replacing Microsoft’s Exchange or SharePoint?

Whether Google makes a success of Chrome OS or not, though, it’s going to show us some things about computing in the future. The network isn’t going to be the computer, I think, but it’s going to be one of the computers, a new kind of partner in a much fuzzier relationship between users and computational tools. In that new relationship, there will also be a lot more to worry about in terms of how each piece integrates with the other pieces, and likely more functional segregation of tasks than administrative segregation. The GUI-versus-application thing is an example. A cloud application, like all applications, will have a network subsystem, an application subsystem, and a database subsystem that serve the user appliance. Some smarts will reside in all of these places, and those smarts will be marshaled in some coordinated way to serve the mission. We’re creating a future that blends SOA principles with principles of GUI design, database design, device design, security, and connectivity. It’s the creation and sustaining of that complex web of stuff that forms the opportunity for the future, and also its challenges.

But there’s more! Part of the Google Chrome OS preview was a comment that at least the prototype netbooks that will be deployed in the extensive pre-release test will be equipped with Verizon wireless services. Google and Verizon, once seemingly irreconcilable enemies, are showing increased coziness. Their net neutrality proposal, which was hated and criticized by everyone, including FCC Chairman Genachowski, looks a heck of a lot like what’s likely to emerge from the FCC’s December 21st public hearing on the topic. It’s all about ecosystems, again.

Yet a pride of lions that eats all its prey species quickly dies off. Google knows that as the OTT giant du jour, it can’t afford to let the problems of disintermediation become critical enough for operators like Verizon to reduce network investment or impose usage pricing with tiers that result in what effectively become taxes on new applications. When I survey users about pricing sensitivity, the results are probably unsurprising at one level. They want unlimited-usage pricing the most. They want low-threshold usage pricing the least. In between, what they’d prefer as an alternative to the latter is application-specific pricing, meaning that they’d like to see any premium charge for usage bundled into the charge (which they or advertisers pay) associated with the application or experience. That way they don’t have to worry about a secret price being added to the visible price and called due later on with their monthly bill.

So it may well be that Google recognizes that the Comcast/Level 3 deal , whatever the rights and wrongs of how it should be characterized might be, is still the right industry answer. Charge for what the user wants, all at once. That means having the content provider collect and settle with the access provider.

What about the kiss Genachowski blew at usage pricing, then?  It may be that he’s simply waving a troll at the kids, creating a threat that makes a spindly carrot look more appetizing.

1  Comment on this Post

There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when other members comment.
  • Webastra
    Rebalancing computing activity between the cloud and the client is really the key to making real-time applications such as VoIP applications work effectively on clouds. Separating call processing from call management and putting call management and reporting on the cloud, while letting the client handle call processing results in calls per second rate that are 8-10 orders of magnitude higher. The application subsystem and the database subsystem can reside in the cloud while the network subsystem gets pushed towards the end devices. This also provides an opportunity to treat the media traffic from end devices with higher QoS without relying on the cloud to do so. If Google Voice has to scale in high call rate environments it will have to deploy a distributed call processing model.
    0 pointsBadges:

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to:

Share this item with your network: