|
Indian scientist creates device that makes net access simpler, faster, cheaper IF DIGITAL divide is all about scarce Internet access, then an Indian professor at Princeton University seems to have the plan for the perfect bridge — a technology that makes net connectivity simpler, cheaper and faster. Vivek Pai and his team have created what they call Hash- Cache, a novel way of caching ( storing frequently accessed web content). They started ‘ caching’ data on a local hard drive rather than using costly bandwidth to retrieve the same data again and again from the net server. Its scope is so immense that the prestigious Technology Review of the Massachusetts Institute of Technology ( MIT) has included it in this year’s list of top 10 technologies. Last year’s top ten list also included a design by Krishna Palem, a professor of computing at Rice University. “ This is going to be the decade of India’s dominance,” said the editor- in- chief Jason Pontin. Pai’s logic is that despite all its dynamism, the bulk of the content in the web does not change too much too often. Still current technologies require large amounts of hard drive space to hold and process data received from the Internet. These data are then indexed by random access memory ( RAM) and stored on the hard drive. That is expensive. “ Though hard disk space is growing more affordable, RAM is still comparatively costly,” said independent programmer Sabu K. J. In many countries, people may not have the money to invest in highbandwidth systems. Besides, RAM requires electricity, which may not be there all the time. HashCache abolishes the index and slashes RAM and power requirement by roughly 90 per cent. “ It can be deployed in a very low- cost box,” Pai said. HashCache achieves this by turning the URL of each stored web “ object” like a chunk of text or an image on a page into a shorter number, using what mathematicians call a ‘ hash function’. While other caching systems use a lot of RAM to store each hash number, HashCache makes the number define the spot on the disk where its corresponding web object can be located. “ By using the hash to directly compute the location, we can get rid of the index entirely,” Pai said. So it will take just enough RAM to do the maths and retrieve the object. HashCache is being field- tested in countries where digital inclusion can mean social and economic change. |