The network of computers which will manage and analyse the deluge of data from the Large Hadron Collider at CERN has been unveiled, three weeks after the first particle beams were sent around the giant physics machine.
The LHC Computing Grid (LGC) will be used to analyse the 15 petabvtes of data created annually by the world's largest physics experiment. The grid spans 33 countries and offers computing power equivalent to 140 computing centres, Cern said.
"The worldwide LHC Computing Grid is a vital pillar of the LHC project," said Jos Engelen, chief scientific officer for the LHC project. "It is an absolute necessity for analysis of the LHC data. It is the result of a silent revolution' in large scale computing over the last five years."
The data collected from the LHC is pumped out over special fibre optic networks to eleven Tier One' computing facilities one of which is Rutherford Appleton Labs in the UK. There, initial processing begins and the data is further distributed to another 140 labs around the globe.
"We can routinely process 250,000 jobs a day," said Ian Bird, leader of the Worldwide LHC Computing Grid project. "And we can achieve peaks of 500,000 jobs without problems." These jobs would take several days on a single processor, hence the need for the combined power of distributed computers.
Despite the initial success of the project, the grid will effectively remain unused until next year, as a helium leak shut down the LHC after just a few weeks. The project also saw trouble from Greek hackers, who attacked the network and took down their website last month.
For more on this massive IT project, read our feature about the UK contribution here.
