Batavia (IL) – The Large Hadron Collider (LHC) may be out of service until next spring, but at least the supercomputer grid that has been designed to sift through the massive amounts of data produced by the LHC is operational. A total of 152 computer centers in 33 countries are part of the Worldwide LHC Computing Grid (WLCG) and provide the combined computational horsepower of 140,000 processing cores.
The WLCG went online earlier this month, roughly three weeks after the after the first particle beams were injected into the LHC. The grid’s center is a “Tier-0” computing site located at CERN in Geneva, Switzerland. All data passes through this central hub, which is claimed to provide slightly less than 30% of the total compute capacity of the grid.
About 70% of computing horsepower is provided by additional tiers: Dedicated optical fiber networks distribute LHC data from CERN in Geneva, Switzerland to eleven major “Tier-1” computer centers in Europe, North America and Asia, including those at DOE’s Brookhaven National Laboratory in New York and Fermi National Accelerator Laboratory in Illinois, which provide about 20% of the computing capacity.
From the Tier-1 centers, data is dispatched to more than 140 “Tier-2” centers in Australia, Belgium, Canada, China, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Hungary, Italy, India, Israel, Japan, Republic of Korea, the Netherlands, Norway, Pakistan, Poland, Portugal, Romania, Russia, Slovenia, Spain, Sweden, Switzerland, Taipei, Turkey, the U.K, Ukraine, and twelve in the U.S. The 140 Tier-2 centers provide about 50% of the entire computing capacity of the grid.
CERN said that raw LHC data will be distributed to the Tier-1 centers, all of which have sufficient storage capacity for a large fraction of the data and are operated around the clock. Analysis tasks requiring access to large subsets of the raw, processed, and simulated data will take place at the Tier-1 centers. The processed data will then be sent to the Tier-2 centers, each consisting of one or several collaborating computing facilities, which can store sufficient data and provide adequate computing power for end-user analysis tasks and Monte Carlo simulation.
Individual scientists will also access these facilities through Tier-3 computing resources, which can consist of local clusters in a University Department or even individual PCs.
CERN that its Tier-0 computing center runs on about 30,000 CPUs and houses about 5 PB (5 million GB) of disk storage and 16 PB of tape storage. Optical 10 Gb/s fiber links connect the computer center to the 11 “Tier-1” sites. There is no exact information what exact hardware is used globally, but CERN noted mentioned that the WLCG’s total computing capacity equals about 140 million SPECint2000. Considering the fact that the average dual-core CPU in use today provides about 2000 SPECint2000, the performance number indicates that the grid provides access to about 140,000 processing cores. The total storage capacity for data is about 60 PB of disk storage and 50 PB of tape storage. The LHC is expected to produce about 15 PB (15 million GB) of data every year. When in operation, the collider generates enough data to fill six CDs every second.
The contribution of the U.S. is provided through the Open Science Grid and consists of two Tier-1 centers and twelve Tier-2 centers in eleven states. 16.7 PB of storage capacity are provided as well.
“As the first point of contact for LHC data in the United States, the computing centers at Brookhaven and Fermilab are responsible for storing and distributing a great amount of this data for use by scientists around the country,” commented said Michael Ernst, director of Brookhaven National Laboratory’s Tier-1 Computing Center, on the launch of the WLCG. We’ve spent years ramping up to this point, and now, we’re excited to help uncover some of the numerous secrets nature is still hiding from us.”
The WLCG went online earlier this month, roughly three weeks after the after the first particle beams were injected into the LHC. The grid’s center is a “Tier-0” computing site located at CERN in Geneva, Switzerland. All data passes through this central hub, which is claimed to provide slightly less than 30% of the total compute capacity of the grid.
About 70% of computing horsepower is provided by additional tiers: Dedicated optical fiber networks distribute LHC data from CERN in Geneva, Switzerland to eleven major “Tier-1” computer centers in Europe, North America and Asia, including those at DOE’s Brookhaven National Laboratory in New York and Fermi National Accelerator Laboratory in Illinois, which provide about 20% of the computing capacity.
From the Tier-1 centers, data is dispatched to more than 140 “Tier-2” centers in Australia, Belgium, Canada, China, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Hungary, Italy, India, Israel, Japan, Republic of Korea, the Netherlands, Norway, Pakistan, Poland, Portugal, Romania, Russia, Slovenia, Spain, Sweden, Switzerland, Taipei, Turkey, the U.K, Ukraine, and twelve in the U.S. The 140 Tier-2 centers provide about 50% of the entire computing capacity of the grid.
CERN said that raw LHC data will be distributed to the Tier-1 centers, all of which have sufficient storage capacity for a large fraction of the data and are operated around the clock. Analysis tasks requiring access to large subsets of the raw, processed, and simulated data will take place at the Tier-1 centers. The processed data will then be sent to the Tier-2 centers, each consisting of one or several collaborating computing facilities, which can store sufficient data and provide adequate computing power for end-user analysis tasks and Monte Carlo simulation.
Individual scientists will also access these facilities through Tier-3 computing resources, which can consist of local clusters in a University Department or even individual PCs.
CERN that its Tier-0 computing center runs on about 30,000 CPUs and houses about 5 PB (5 million GB) of disk storage and 16 PB of tape storage. Optical 10 Gb/s fiber links connect the computer center to the 11 “Tier-1” sites. There is no exact information what exact hardware is used globally, but CERN noted mentioned that the WLCG’s total computing capacity equals about 140 million SPECint2000. Considering the fact that the average dual-core CPU in use today provides about 2000 SPECint2000, the performance number indicates that the grid provides access to about 140,000 processing cores. The total storage capacity for data is about 60 PB of disk storage and 50 PB of tape storage. The LHC is expected to produce about 15 PB (15 million GB) of data every year. When in operation, the collider generates enough data to fill six CDs every second.
The contribution of the U.S. is provided through the Open Science Grid and consists of two Tier-1 centers and twelve Tier-2 centers in eleven states. 16.7 PB of storage capacity are provided as well.
“As the first point of contact for LHC data in the United States, the computing centers at Brookhaven and Fermilab are responsible for storing and distributing a great amount of this data for use by scientists around the country,” commented said Michael Ernst, director of Brookhaven National Laboratory’s Tier-1 Computing Center, on the launch of the WLCG. We’ve spent years ramping up to this point, and now, we’re excited to help uncover some of the numerous secrets nature is still hiding from us.”