Feature: The grid that sifts for dark matter

Cryogenic Dark Matter Search detectors. The CDMS experiment uses five towers of six detectors each. Photo credit: Reidar Hahn.

Think of grid computing as a sieve that physicists use to sift out those rare events that might just be signs of dark matter — the mysterious substance that appears to exert gravitational pull on visible matter, accelerating the rotation of galaxies.

FermiGrid, the campus grid of Fermilab and the interface to the Open Science Grid, recently helped researchers from the Cryogenic Dark Matter Search experiment do just that: identify two possible hints of dark matter.

Dark matter has never been detected. And although the CDMS team cannot yet claim to have detected it, their findings have generated considerable excitement in the scientific community.

“This is a very intriguing result,” said Lauren Hsu, a CDMS researcher at Fermilab who announced the experiment’s results at a talk last December. “As an experimenter you always wish you had more data. I’m really interested to see what our next results will be.”

The experiment, managed by Fermilab and bringing together scientists from several universities, operates a set of detectors in the Soudan Mine in Minnesota, a half-mile underground.

The CDMS collaboration processed eight terabytes of data from these detectors for this result, more than twice as much as all their previous data combined. By using FermiGrid resources, the collaboration accelerated its data processing by about a factor of ten, according to Hsu.

“The total ‘human’ time to process the dataset was one month,” she said.

“Our accounting shows that CDMS used 173,300 wall-clock hours on FermiGrid in 2009,” said Steve Timm of Fermilab’s Computing Division. “That’s the equivalent of using 18 CPUs for a solid year. Of course, the CDMS usage actually came in much more concentrated bursts than that.”

Timm said that having a large portion of their dataset available on the BlueArc network-attached storage appliance for easy access was a key factor in the performance.

Given the statistical nature of the research, the collaboration will need more data to identify dark matter definitively.

“While (the CDMS) result is consistent with dark matter, it is also consistent with backgrounds,” said Fermilab Director Pier Oddone, referring to signals that can be attributed to other, known sources. In 2010, the collaboration will install an upgraded detector expected to produce about three times the amount of data. The researchers will then be able to confirm whether they have indeed found dark matter particles.

-Anne Heavey, Fermilab with materials contributed by Fermilab Today. To learn more about these results, watch the talk that was given at Fermilab about CDMS results.