"Big" think

Interesting comments from the Chicago folks. The Cloud Computing model has all of the brains with the data, and lightweight display units, not unlike Larry Ellison's thin termials of last century.

While we know that thin terminals failed last time, they were the model for IT in the 60s and 70s, with mainframes and later Vaxes. So perhaps they could come back.

Clearly today's model of highly complex and unreliable power on the desktop is unsustainable.

Back when it was called the "Information Superhighway" there were visions of today's 'net, but also one radically different one that had some legs. William H Murray talked about a big, massively interconnected mesh network of fiber optic speeds, with all the brains at the edges.

The dumb center, smart edges, model has not happened but its not clear that it is not at least as viable today as it was in 1992.

Its not clear that your payroll, tax, or inventory application needs to be run on Google's cloud. They are not complex, and could be run on a typical iPod nano's CPU.

In the era of free silicon, its not clear to me how much computing needs to be done in the "cloud computing" and how much can be done by my wristwatch.

Along with the privacy issues, think of the possible legal extensions as far as trade and export control laws.

I had the opportunity recently to talk with some trade law experts. Being Canadian, Canada has some different relationships with other nations of the world than the US has (think Cuba)

If as a Canadian Company, I have a Cuban client, but my product or service is in a cloud out of California. What are the trade or export implications?

I completely agree with fishtoprecords about us being in an 'Era of Free Silicon'. The last time I bought HP servers they gave us the second [4 core] CPU free!

In my world, the main "cloud" services are CDNs [they host other people's media across over globally distributed fast servers] which exist because "backbone" bandwidth is expensive and h/ware is cheap.

Interesting thought experiments :- would S3 function better if it were distributed over 1000s of nodes? My data just gets stored in whichever 2 nodes are nearest to me, then migrated out further if it's requested from new locations.

If you ran a global website (think eBay) which needed to have a single authoritative database, would you end up building an entire global fibre network just connecting to that DB master server? Could there come to be some giant warehouse in San Diego that had all the key live databases of the world?

Going to be interesting to see how the economics of all this actually work out...

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

URL:
Remember me? YesNo

Comments:

Rough Type is:

Written and published by
Nicholas Carr

Designed by

JavaScript must be enabled to display this email address.

What?