The CERN Tape Archive Beyond CERN An Open Source Data Archival System for HEP

• Published in: EPJ Web of conferences 2024, Vol.295, p.1048
• Main Authors: Davis, Michael, Afonso, João, Bachmann, Richard, Bahyl, Vladimír, Camarero Vera, Jorge, Leduc, Julien, Oliver Cortés, Pablo, Rademakers, Fons, Wardenær, Lasse, Yurchenko, Volodymyr
• Format: Article
• Language: English
• Subjects: Archives & records; CERN; Computational grids; Data storage
• ISSN: 2100-014X; 2101-6275; 2100-014X
• DOI: 10.1051/epjconf/202429501048

The CERN Tape Archive (CTA) is the successor to CASTOR and the tape backend to EOS. It was designed to meet the needs of archival storage of data from LHC Run–3 and other experimental programmes at CERN. In the wider Worldwide LHC Computing Grid (WLCG), the tape software landscape is quite heterogeneous, but we are entering a period of consolidation. A number of sites have reevaluated their options and are choosing CTA for their future tape archival storage needs. However, CTA’s original mandate imposed several design constraints which are not necessarily optimal for external sites. In this contribution, we show how CERN has engaged with the wider HEP community and collaborated on improvements which allow CTA to be adopted more widely. We detail community contributions to allow CTA to be used as the tape backend for dCache; to facilitate migrations from other tape systems such as OSM and Enstore; and improvements to CTA building and packaging to remove CERN-specific dependencies and to allow easy distribution to external sites. Finally, we present a roadmap for the community edition of CTA.