Harnessing The Awesome Power Of Smartphones

The processing power of latest generation of smartphones is quite awe inspiring. With quad core microprocessors clocked at over 1 GHz, dedicated graphics processors and system on chip (SoC) architectures they are far more powerful than the average desktop computer of just a few years ago, and a typical 1980s supercomputer. And it all fits neatly in your pocket.

The problem is that for the most part that is where it stays. All that amazing computer power for most of the time is just going to waste. And that is just your phone. Think of all that computing power which is sitting in pockets and handbags everywhere- just how many potential supercomputers are being wasted?

Image Source: Creative Common Flicker  Image Credit

However there are plans afoot to change all that with cluster computing. This intends to combine all the potential computing power that is currently lying dormant and directing it is such computer intensive projects such as climate modelling.

Cluster computing has been used in numerous projects involving desktop computers; examples are SETI, the search for extra-terrestrial intelligence and LHC Test for Theory, which analyses data from the LHC at CERN in the search for new fundamental particles, both of which harness computer power when the desktops are idle.

The problem with smartphone clusters is battery consumption; smartphone batteries have short lifetimes, especially when using their full computing power. But, this is not a problem when phones are being charged, and many smartphone users charge their phones overnight.

In an experiment, six low end Android phones were connected in a network, and between them they were able to carry out around 30 million calculations a second (30 megaflops). High end Android phones can carry out around 100 megaflops so, allowing for losses, a similar network would be capable of around 500 megaflops. Just imagine the potential computing power of several hundred networked smartphones all communicating over Wi-Fi; this could be the case on any train journey for instance.

Essentially a network of smartphones forms a mobile cloud that users can join and leave as they wish; the more users, the greater the computing power that is available. Currently a number of phone clustering business models are being examined. In order to encourage phone users to join a cluster, attractive inducements such as subsidised pay as you go phones are being considered. Owners of pay monthly phones could also be encouraged to participate by emphasising the social aspects of clustered computing, for instance through Facebook groups.

This is a guest post by Claire Chat a Londoner interested in technology in general as well as in the mobile and telecommunication industry. If you want Claire to write you specific content, you can find email her here or contact her on Twitter (Claire_Chat).