|
Peter Cochrane's Hard Drive 1999 Tapping into the power of many THROUGHOUT my career I have run into problems beyond the computing power available. While I have had the power to solve 95 per cent of my problems, there has always been an annoying five per cent that demanded some distant supercomputer well beyond financial or practical realisation. Despite the exponentially growing capabilities of the PC, there are always engineering and scientific problems that require even more. In recent years I have seen people tackle immense problems associated with turbulent systems in aeronautical design, many-body problems, network modelling and prediction with very impressive results. But there is now an alternative to the supercomputer - the use of hundreds or thousands of PCs in parallel. To realise such a solution the problem itself has to be amenable to being broken up into small tasks - or decimation. In 1994 Nasa built a parallel processing computer of 16 486 processors to establish the proof of principle. An American government programme followed in 1996 with 9,000 linked Pentium processors realising the world's first teraflop machine - ie 1,000 billion floating point operations a second. In both cases these machines had all the chips in one giant machine. Since then there have been several extensions to even bigger arrays and even more teraflops, but perhaps more exciting has been the advent of the distributed supercomputer with PCs linked via the Internet. One area that lends itself to this sort of treatment is code-cracking. Instead of using a single massive number cruncher to try to break a code, thousands of PCs can be used in parallel. Each can be assigned a portion of the task and let loose to crunch through a range of possibilities. In my own company we regularly commandeer thousands of machines outside and during the working day to release their down time across our intranet. Go home and your machine becomes part of a giant computational engine for the night. Of course, you need the agreement of the donor so you can load the software and access the spare PC's power and time. Given the success of this approach in the professional domain, I went looking for some worthy cause to take up the slack of my personal computing power on my laptop and at home. I soon stumbled across the Seti (Search for Extra Terrestrial Intelligence) programme linked into the great radio telescope at Arecibo, Puerto Rico. This was originally a fully funded Nasa programme that fell foul of funding cuts in 1997. In the true spirit of scientific exploration, the Seti team managed to stagger on by recruiting a rapidly expanding group of people willing to dedicate their PC downtime to the search for ET. For me Seti has always seemed to be both a real long shot and a worthy cause. So I decided to join the programme for two key reasons. First, you get a screen full of the radio spectrum being processed that makes a really neat screen saver and talking point. Second, it might just be my laptop that discovers ET and sets all the bells ringing. If it does - I want to be there. If you care to join me, check out setiathome.ssl.berkeley.edu/, and log into the Telegraph Connected Group section. Peter Cochrane holds the Collier Chair for the Public Understanding of Science & Technology at the University of Bristol. His home page is: |
Telegraph Group Limited endeavours to ensure that the information is correct but does not accept any liability for error or omission.
Users are permitted to copy some material for their personal use, but may not republish any substantial part of the data either on another website or as part of any commercial service without the prior written permission of Telegraph Group Limited.