In the framework of WLCG, Tier-1s need to manage large volumes of data ranging in the PB scale. Moreover they need to be able to transfer data, from CERN and with the other centres (both Tier-1s and Tier-2s) with a sustained throughput of the order of hundreds of MB/s over the WAN offering at the same time a fast and reliable access also to the computing farm. In order to cope with these challenging requirements, at INFN Tier-1 we have adopted a storage model based on StoRM/GPFS/TSM for the D1T0 and D1T1 Storage Classes and on CASTOR for the D0T1. In this paper we present the results of the commissioning tests of this system for the ATLAS experiment reproducing the real production case with a full matrix transfer from the Tier-0 and with all the other involved centres. Noticeably also the new approach of direct file access from farm to data is covered showing positive results. GPFS/StoRM has also been sucessfully deployed, configured and commissioned as storage solution for an ATLAS INFN Tier-2, specifically the one of Milano. The results are shown and discussed in this paper together with the ones obtained for the Tier-1.

Commissioning of a StoRM based data management system for ATLAS at INFN sites / A. Brunengo, C. Ciocca, M. Corosu, M. Pistolese, F. Prelz, L. Rinaldi, E. Ronchieri, V. Sapunenko, A. Andreazza, S. Barberis, G. Carlino, A. Cavalli, S. Dal Pra, L. Dell'Agnello, D. Gregori, B. Martelli, L. Perini, A. Prosperini, P. Ricci, D. Vitlacil. - In: JOURNAL OF PHYSICS. CONFERENCE SERIES. - ISSN 1742-6588. - 219:6(2010), pp. 062042.062042.1-062042.062042.8. ((Intervento presentato al 17. convegno CHEP : International conference on computing in high energy and nuclear physics tenutosi a Praga nel 2009 [10.1088/1742-6596/219/6/062042].

Commissioning of a StoRM based data management system for ATLAS at INFN sites

A. Andreazza;L. Perini;
2010

Abstract

In the framework of WLCG, Tier-1s need to manage large volumes of data ranging in the PB scale. Moreover they need to be able to transfer data, from CERN and with the other centres (both Tier-1s and Tier-2s) with a sustained throughput of the order of hundreds of MB/s over the WAN offering at the same time a fast and reliable access also to the computing farm. In order to cope with these challenging requirements, at INFN Tier-1 we have adopted a storage model based on StoRM/GPFS/TSM for the D1T0 and D1T1 Storage Classes and on CASTOR for the D0T1. In this paper we present the results of the commissioning tests of this system for the ATLAS experiment reproducing the real production case with a full matrix transfer from the Tier-0 and with all the other involved centres. Noticeably also the new approach of direct file access from farm to data is covered showing positive results. GPFS/StoRM has also been sucessfully deployed, configured and commissioned as storage solution for an ATLAS INFN Tier-2, specifically the one of Milano. The results are shown and discussed in this paper together with the ones obtained for the Tier-1.
distributed computing ; GRID ; LHC ; ATLAS ; high energy physics
Settore FIS/04 - Fisica Nucleare e Subnucleare
Settore ING-INF/05 - Sistemi di Elaborazione delle Informazioni
2010
Article (author)
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/139986
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact