Project

Current high energy physics experiments such as CMS aim to explore new territories where new physics is expected, and where a huge amount of data needs to be collected and analyzed. The accomplishment of these scientific projects require computing resources beyond the capabilities of a single user or group, thus the data is treated under the grid infrastructure according to the Worldwide LHC Computing Grid [1]. The high energy physics group has a long tradition in computing. Taking profit from the gained experience, since early 2002 the group contributes to the computing tasks at the CMS experiment. In 2005 IFCA and CIEMAT established the CMS Spanish Tier-2, and since then provide service to the whole collaboration. A Tier-2 stores sufficient data and provides adequate computing power for specific physics analysis tasks. They handle analysis requirements and proportional share of simulated event production and reconstruction. By March 2012, IFCA counts with over 2000 CPUs (shared with the GRID-CSIC project) and more than 2000 terabytes of disk space.  Some numbers about the performance of the IFCA site during 2010: data was transferred at impressive rates, witnessing peaks of 80 megabytes per second sustained during a day, the equivalent of one full DVD of data a minute. Around a thousand of physicists from all around the world run their computing jobs at IFCA, the site reached peaks of almost 30,000 jobs per day. The site usage is constantly reported through several CMS tools [2], providing up to date information about job statuses and accounting information. Continuous hardware and software upgrades need to be deployed in order to adapt to the CMS requirements, mainly driven by the amount of data to be produced and analysed.  Permanent monitoring of the site guarantees the high quality operational status of the site. This allows the early detection of any eventual issue and the prompt actions to solve them. All those tasks are carried by experienced Grid and CMS support team at IFCA. Moreover, the group is deeply involved in the development of tools that improve the efficiency and friendliness of the analysis user applications, thus allowing a better use of the resources.  

[1]  http://lcg.web.cern.ch/LCG/public/tiers.htm

[2]  http://dashboard.cern.ch/cms/