Global grid for Big Bang research reaches milestone

One major test remains as grid project prepares to go live next year

A huge 100,000-PC grid-computing network being built to help research the origin of the universe passed the third of four major tests recently when it reached a data-transfer milestone, with up to 1GB/sec. of physics data sent over the global grid.

In an announcement yesterday, the European Organization for Nuclear Research (CERN) said the data was transferred from CERN in Geneva to 12 major computer centers around the globe. More than 20 computing facilities at universities and colleges were also used in the testing. The data transfers were made to analyze real-time storage, distribution and analysis of the data while the grid is being built and refined.

The grid project is being assembled to help deliver physics data around the world from the Large Hadron Collider (LHC), a giant particle collider that will be used by researchers to help learn more about the Big Bang theory of the creation of the universe. The LHC is scheduled to begin operating at CERN next year (See story "Planet-scale grid").

Francois Grey, an IT spokesman for CERN, said the latest test was part of a scheduled series of four tests that must be completed before the grid goes live next year.

"This was the first time that this model was tested with real data" at the 1GB/sec. data-transfer rate, Grey said. A previous grid test conducted last April involved seven major computer centers in the U.S. and Europe and transferred data at lower speeds -- sustained rates of up to 600MB/sec.

The 100,000-PC grid will help CERN and a worldwide network of researchers and scientists harness a huge amount of computing power and storage capacity that it couldn't build on its own, Grey said.

"We have absolutely no means, budgetary or otherwise, to assemble that," he said. "So the computing power is definitely something that needs to be spread out."

The grid will also be able to store some 15 petabytes of data -- 15 million gigabytes -- annually, and will have to be continually expanded because the data will have to be continuously preserved for analysis. "That's a big challenge, and that's why we need to distribute the storage" over the worldwide grid, Grey said.

Next year, the LHC particle collider will be used by scientists to begin smashing protons and ions together in a massive, multinational experiment to understand what the universe looked like tiny fractions of a second after the Big Bang. The particle accelerator will release a vast flood of data on a scale unlike anything seen before, which is why the grid computing network is needed. The LHC, which is being built near Geneva, will be a circular structure 17 miles in circumference and will eventually produce data at up to 1.8GB/sec.

The grid will send the data from CERN over the Internet and over specialized high-speed links around the world, Grey said, via a middleware software stack that can be used by researchers. When a researcher accesses the grid for data analysis, the software can automatically determine the closest source of the data in the grid and the closest source of computing power, and it can automatically perform authentications, record keeping and other related tasks, all in a secure environment, he said.

The final major test of the grid is scheduled for this summer.

Once the testing is completed, four experiments will produce data that will be sent to the grid from the LHC. They are Atlas ATLAS, CMS (Compact Muon Solenoid), ALICE (A Large Ion collider Experiment) and LHCb.

The experiments involve huge underground detectors that will identify collisions of particles, providing data that can be produced in large images, like a fireworks display, Grey said. The experiments can help researchers to learn whether other particles were formed only briefly when the universe was formed during the Big Bang and to find out what made up those missing particles.

"It's detective work to figure out what was there and how it acted," Grey said. "It's real, fundamental science. This grid is being built to handle huge amounts of data that is being spewed out. There's Nobel Prizes at stake, so it's pretty exciting."

CERN, which is a European laboratory for particle physics, is leading the LHC and grid efforts. In the U.S., the projects include grid participation from the Brookhaven National Laboratory, the Fermi National Accelerator Laboratory, Purdue University, California Institute of Technology, the University of California at San Diego, the University of Wisconsin-Madison, the University of Florida and the University of Nebraska-Lincoln.

Copyright © 2006 IDG Communications, Inc.

Shop Tech Products at Amazon