Particle physics' current flagship LHC, the world's most powerful accelerator in Geneva, generates data amounting to around 15 petabytes (15 million gigabytes) each year, enough to fill hundreds of thousands of DVDs a year. Thousands of scientists around the world analyse this flood of data – an enormous challenge with respect to data storage and computing power.

Worldwide computer network

To meet this challenge, the LHC designers have turned to the concept of grid computing, in which computers distributed around the globe are linked together in such a way that they can be used as a powerful supercomputer by users from all over the world. The data from the experiments are thus no longer stored and processed in just one place, but distributed to a series of computing centres with sufficient storage capacity, from where they are transferred to other facilities and ultimately to the participating scientists. DESY is contributing large computing and storage systems to this worldwide computer grid.

Tier-2 centre at DESY

Computing centres around the globe are cooperating with CERN to establish this gigantic computer network called the LHC Computing Grid (LCG). The grid comprises various levels, called tiers. The raw data generated by the LHC are first backed up on tape at CERN in Geneva, the Tier-0 centre. Following initial processing, these data are then distributed to 11 international Tier-1 centres for storage and reconstruction. These centres have the requisite storage capacity for large data volumes and are constantly connected to one another via the grid. One of these Tier-1 centres is the Grid Computing Centre Karlsruhe (GridKa) of the Research Centre Karlsruhe, which constitutes one of the primary central European nodes of the LCG.

The Tier-1 centres in turn make the data available to the Tier-2 facilities further down in the grid hierarchy. These facilities, which comprise either a single computing centre or an interconnected number of them, constitute the decisive level for the scientific analysis of the data using grid tools. DESY operates such a Tier-2 centre distributed across its Hamburg and Zeuthen sites, providing large computing and storage systems for the LHC experiments ATLAS, CMS and LHCb. It is one of the largest of the more than 140 centres distributed around the globe and is used via the grid by research groups all over the world. The scientists access the LCG computing network via local computing centres at their institutions (Tier-3) or from the computers at their desks (Tier-4). This enables scientists all over the world to evaluate and analyse the stored data using the computing power of the LCG.