The Next Big Thing
Posts about next generation technologies and their effect on business.

A Different Kind of Cloud Computing

In reading through the recent blog on cloud computing, my mind started wandering to this very topic. However, I was not thinking about cloud computing in the "traditional" sense of compute-as-a-service, but in a more controversial fashion.

 

Let's consider the number of mobile devices across the globe, to be estimated at 2.2 billion (according to the CIA's The World Factbook). Assume that the average mobile phone has 256 MB of memory (conservative estimate, since the new iPhone comes with 8 or 16 GB of flash memory), with 128 MB free for use. That provides us with a data center consisting of over 2 billion processors and more than 250 Petabyte (PB). To put that into comparison: Google processes a "mere" 20 PB of information per day according to Niall Kennedy's Blog.

 

The traditional http://setiathome.berkeley.edu/ project has almost 5.5 million users signed up, contributing more than 21 billion CPU hours. These compute resources are dependent on the devices to be "connected". Since the mobile phones are "always on", the "data center in the cloud" concept can take advantage of this type of capability.

 

However, these types of resources, if harnessed, can provide compute capabilities that are an order of a magnitude larger than the ones currently encountered in the "cloud" computing area. These mobile phones may one day become the data center in the cloud.

Comments
Anonymous | ‎04-10-2008 10:54 PM

You could take this idea a stage further.  One of the ideas I have come across is that mmobile devices offer each other services in a dynamic fashion so for example a device with a high capacity link that its not using fully could share some of the capacity with another nearby device on a bluetooth link for example (we will ignore the challenge of security and billing for the mmoment!).  This raises the thought that rather than the SETI model where a task is loaded onto a processor that then reports back later one might envisage taks that leap from device to device using some local link.  Truly distributed!

Anonymous | ‎04-11-2008 07:29 PM

I agree, mesh networks will play a tremendous role in this type of cloud computing setup: self-governing networks that adapt according to need and network resource availability.  

| ‎07-01-2009 10:26 PM

You add onto this hive computing approach the contextual knowledge of the device and you can tackle interesting traffic flow problems as well as have a computational network.

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the community guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Search
Showing results for 
Search instead for 
Do you mean 
About the Author
Featured
Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation.