INFN-CNAF, the italian Tier-1, houses more than 10 PB of tape space and 6 PB of disk space used by the global physics user communities including the LHC experiments. In this context, an innovative solution to provide the highest level of reliability, scalability and performances on the installed mass storage system has been developed and used on the production system. The solution, called Grid Enabled Mass Storage System (GEMSS), is based on two components: a layer between the IBM GPFS and TSM HSM, and StoRM, designed to allow direct access to the storage resources by file protocol as well as standard Grid protocols. During this year, both LHC and non-LHC experiments have agreed to use GEMSS with SRM endpoint at INFN Tier-1 after the excellent results observed during their GEMSS validation phase. In this paper we describe the storage system setup, the data workflow and computing activities of the LHC experiments, and main results obtained at INFN-CNAF Tier-1 in the first year of LHC data taking. In particular, we show the system response to a sustained huge load on the mass storage system and the computing infrastructure.
INFN-CNAF Tier-1 Storage and Data Management Systems for the LHC Experiments / D Andreotti;S Antonelli;A Cavalli;C Ciocca;S Dal Pra;L dell'Agnello;C Genta;D Gregori;B Martelli;A Prosperini;P P Ricci;L Rinaldi;V Sapunenko;V Vagnoni. - In: JOURNAL OF PHYSICS. CONFERENCE SERIES. - ISSN 1742-6596. - STAMPA. - 331:(2011), pp. 052005-052011. [10.1088/1742-6596/331/5/052005]
INFN-CNAF Tier-1 Storage and Data Management Systems for the LHC Experiments
RINALDI, LORENZO;
2011
Abstract
INFN-CNAF, the italian Tier-1, houses more than 10 PB of tape space and 6 PB of disk space used by the global physics user communities including the LHC experiments. In this context, an innovative solution to provide the highest level of reliability, scalability and performances on the installed mass storage system has been developed and used on the production system. The solution, called Grid Enabled Mass Storage System (GEMSS), is based on two components: a layer between the IBM GPFS and TSM HSM, and StoRM, designed to allow direct access to the storage resources by file protocol as well as standard Grid protocols. During this year, both LHC and non-LHC experiments have agreed to use GEMSS with SRM endpoint at INFN Tier-1 after the excellent results observed during their GEMSS validation phase. In this paper we describe the storage system setup, the data workflow and computing activities of the LHC experiments, and main results obtained at INFN-CNAF Tier-1 in the first year of LHC data taking. In particular, we show the system response to a sustained huge load on the mass storage system and the computing infrastructure.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.