If you ever doubt the importance of parallel processing think again. Cast your eyes back across the 2,000 year span of Western history and what common denominator, from then until now, are you likely to see? There is one that leaps to my mind; the high value that is always placed on money.
Money or the search for profit is the one thing that often drives change. And money talks. The parallel processing techniques of next-generation sequencing have caused the cost of DNA sequencing to drop significantly. Research from the National Human Genome Research Unit shows that costs plummeted from around $1,000 in October 2007 to $1 in October 2009, outstripping the exponential curves of Moore’s Law.
The point is that next-generation sequencing, based on parallel processing, has become a viable research option for increasing numbers of researchers. As a result, the amount of DNA data collected has mushroomed over the last few years and the more data, the more chance of making dramatic DNA breakthroughs.
It’s the parallel processing techniques that have dramatically slashed the cost of DNA research, which in turn is leading to more research, more knowledge and ultimately more breakthroughs.
And parallel processing is also at the centre of advancements in energy-efficient car design. Cambridge University Eco Racing (CUER) is one of a swathe of university and technical teams developing a solar-powered car for the 2011 World Solar Challenge.
This 3,000km gritty jaunt through the harsh Australian outback will depend solely on harnessing the power of the sun. And it’s not just a jolly jape. Data collected from the cars will inform the commercial development of energy-efficient vehicles.
CUER uses parallel programming to divide a mesh of millions of small cells into several parts to calculate the fluid flow across the boundaries of cells and cell temperature and pressure. This informs the actual design of the car.
Within the larger context of climatic and economic factors the development of energy efficient cars is going to become big business in the next 20 years. Access to dwindling oil supplies is becoming a broad concern, climate change scientists predict extreme weather and companies that operate in sectors that are affected are busily working to reposition themselves so they can meet the changes head on.
All of this means that parallel processing, albeit invisibly and quietly to most people, is going to become ever more important. Despite the complexities, headaches and coding spats, parallel processing is an exciting place to be and I’m betting that before long it will be the only game in town.