Disk storage at CERN: Handling LHC data and beyond

20th International Conference on Computing in High Energy and Nuclear Physics

物理学;计算机科学

Espinal, X.^1 ; Adde, G.^1 ; Chan, B.^1 ; Iven, J.^1 ; Presti, G. Lo^1 ; Lamanna, M.^1 ; Mascetti, L.^1 ; Pace, A.^1 ; Peters, A.^1 ; Ponce, S.^1 ; Sindrilaru, E.^1

CERN European Laboratory for Particle Physics, Genève

1211, Switzerland^1

关键词: Data intensive;    Highly reliable;    LHC experiments;    Multi-replicas;    Physics experiments;    Scale analysis;    Storage services;    Usage patterns;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/513/4/042017/pdf
DOI  :  10.1088/1742-6596/513/4/042017

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

The CERN-IT Data Storage and Services (DSS) group stores and provides access to data coming from the LHC and other physics experiments. We implement specialised storage services to provide tools for optimal data management, based on the evolution of data volumes, the available technologies and the observed experiment and users' usage patterns. Our current solutions are CASTOR, for highly-reliable tape-backed storage for heavy-duty Tier-0 workflows, and EOS, for disk-only storage for full-scale analysis activities. CASTOR is evolving towards a simplified disk layer in front of the tape robotics, focusing on recording the primary data from the detectors. EOS is now a well-established storage service used intensively by the four big LHC experiments. Its conceptual design based on multi-replica and in-memory namespace, makes it the perfect system for data intensive workflows. The LHC-Long Shutdown 1 (LSI) presents a window of opportunity to shape up both of our storage services and validate against the ongoing analysis activity in order to successfully face the new LHC data taking period in 2015. In this paper, the current state and foreseen evolutions of CASTOR and EOS will be presented together with a study about the reliability of our systems.