- News Home
17 April 2014 12:48 pm ,
Vol. 344 ,
Using the two high-quality genomes that exist for Neandertals and Denisovans, researchers find clues to gene activity...
A new report from the Intergovernmental Panel on Climate Change (IPCC) concludes that humanity has done little to slow...
Astronomers have discovered an Earth-sized planet in the habitable zone of a red dwarf—a star cooler than the sun—500...
Three years ago, Jennifer Francis of Rutgers University proposed that a warming Arctic was altering the behavior of the...
Officials last week revealed that the U.S. contribution to ITER could cost $3.9 billion by 2034—roughly four times the...
An experimental hepatitis B drug that looked safe in animal trials tragically killed five of 15 patients in 1993. Now,...
- 17 April 2014 12:48 pm , Vol. 344 , #6181
- About Us
Exascale Computing Off to Slow Start in DOE Budget
13 February 2012 9:55 pm
Just as each generation of desktop and laptop computers becomes a bit sleeker and faster than its predecessor, supercomputers have improved steadily for decades. To stay on that track, supercomputer makers are hoping to build their first "exascale" machines between 2018 and 2020. To build those machines, which are 100 times more powerful than today's top performers, researchers need to come up with a host of breakthroughs in everything from making energy-efficient chips to new algorithms for programming them. So supercomputer scientists were watching the release of the Obama Administration's budget closely to see if it contained new pots of money for exascale science. Observers came away both pleased and disappointed.
In December 2011, Congress settled on giving exascale computing $90 million, via the Department of Energy, for the 2012 budget. For the next fiscal year, the Administration today requested only $68 million in DOE funding for exascale science, which includes everything from hardware development, novel mathematical methods, and scientific applications, such as climate modeling software programs. And while that sounds like a disappointing 25% cut, the Administration requested an additional $21 million for DOE for research relating to "data-intensive sciences." This latter category includes research on how to manage the torrents of data emerging from climate models and other advanced supercomputing programs. So between the two programs, the exascale budget request is essentially flat.
"I was hoping to see more of a ramp up," says John Shalf, a supercomputing expert at Lawrence Berkeley National Laboratory in California. But with the 2012 budget now set, and at least a first promise that the program will continue in 2013, Shalf says it's an important start. "There is enough money in FY13 that they can start projects up in 2012 and not have to cancel them in 2013."
Then again, even with the tight fiscal times, Congress may choose to add money to the exascale initiative when it gets involved. The United States is already lagging behind Japan and China when it comes to the world's most powerful supercomputers. And those countries, along with the European Union, Russia, and India, have all underscored their desire to be first to the exascale, in hopes of giving their homegrown computer industries a leg up on the competition. That challenge prompted Congress to back the exascale initiative for 2012, and may encourage them to boost the funds down the road.