Distributed computing

Posted at: Tuesday 29th June 2004 by Ian Betteridge

You might just use your bog-standard 2.4GHz PC for writing inane messages to your friends, but it could become the most powerful computer in the world, or at least part of it. Ian Betteridge explains how

So how can you maximise your contribution to a distributed project, while minimising the impact on the environment? The first option is simply to ensure that you don't leave your machine on all the time, but get maximum performance out of it while it's on. Yes, this is a great excuse to dig into the toolbox and tweak your machine to get the maximum performance, and have a positive impact on the world to boot.

Secondly, you can use a laptop rather than a desktop machine. On average, a laptop consumes around a third of the power of an equivalent desktop, thanks to its low-power components and power management options. If you're using a desktop, then you can also turn off the monitor - which uses the vast majority of the power - whenever possible, and also invest in a low-power TFT panel, which is better for your eyes and consumes fewer resources than a CRT.

Finally, if you can, switch to an electricity company that generates its power through renewable methods, such as wind and wave power, rather than fossil fuels, which will significantly reduce the carbon dioxide emissions.

Several companies, including nPower and Scottish Power, offer schemes where all the electricity you buy comes from renewable sources, and you may even save money. You can check which services are available to you at www.ukgreenpower.co.uk

What difference can you make?
Given that distributed computing projects can achieve the kind of performance that would otherwise require millions of dollars, you might wonder how much difference your machine can make, even if you've got a setup that's optimised for distributed computing. After all, even smaller distributed computing projects achieve performance that's rated in teraflops, compared to the mere gigaflop throughput of even the fastest desktop PCs.

Yet without each individual computer, distributed projects would be nothing. As Vijay Pande puts it: 'A very rough analogy is that of snowflakes and snow. One snowflake can't really achieve much, but combined, they can be really powerful. Similarly, one CPU can't achieve much - but 100,000 can have a very powerful impact.' David Anderson from SETI@home, meanwhile, puts it even more simply: 'We currently have around 500,000 users at any one time. Without the contributions of these individuals, there would be no computing power, and no SETI@home.'

Also, in many cases, these projects are looking for a specific result and any one of the computers in the project could actually find it. According to Keith Davies from the Find-A-Drug project, 'it's rather like a lottery. No one knows in advance who will find a potential drug molecule, although the members who contribute the most computer time obviously have the best chance of finding an active molecule. A number of members also contribute because they know individuals who have been affected by cancer, HIV or other diseases we target -it's easy to do and probably more useful than putting a coin in a collection box'.