Now you know Fridays: Computer Science Olympics
April 27, 2012
Our new Now You Know Fridays series scours the Web for the most interesting news in tech to add to your digital biz know-how!
Computer scientists can be competitive. In fact, there is even an annual contest to see who can sort the greatest amount of digital data in the shortest time. The competition is called the Sort Benchmark and it attracts brainiacs from university and industry computer labs around the world. It’s sort of like a digital Olympics where they can flex their mental muscle in a bid to create the fastest data centre designs.
To understand how this affects everyone’s lives, know this: such sorting prowess promises to be key to coming scientific advances. TechNewsWorld reports that IBM is partnering with Dutch scientists to sort data from the world’s biggest and most sensitive radio telescope.
Not only that, making data sort faster has important implications for anyone who has ever plugged a term into a search engine, looked for book or music recommendations online, or tagged a friend on Facebook. How? Because with every increase in sorting power, comes a decrease in search times. And since the idea behind the Sort Benchmark is to use as few resources as possible, the winning designs can also save money.
As Forbes puts it, tech giants such as Google, Yahoo and Twitter “have been able to change consumers’ lives so profoundly only because they’ve built such sophisticated, back-end computing and data-analytic environments that can quickly process and analyze stunning volumes of data: searches, tweets and other content running through their sites each day.”
In 2010, UC San Diego computer scientists broke the terabyte barrier by sorting more than one terabyte of data in just 60 seconds (that’s 1,000 gigabytes or 1 million megabytes). They also sorted one trillion data records in 172 minutes – tying the previous record, but doing it with four times fewer computing resources.
“If a major corporation wants to run a query across all of their page views or products sold, that can require a sort across a multi-petabyte dataset and one that is growing by many gigabytes every day,” said UC San Diego computer science professor Amin Vahdat, who led the project. “Companies are pushing the limit on how much data they can sort, and how fast.”
According to The Future of Things, the UC San Diego record is especially significant because the technology can be applied to almost all platforms.
Talk about going for gold.