Tuning grid storage resources for LHC data analysis
暂无分享,去创建一个
Grid Storage Resource Management (SRM) and local file-system solutions are facing significant challenges to support efficient analysis of the data now being produced at the Large Hadron Collider (LHC). We compare the performance of different storage technologies at UK grid sites examining the effects of tuning and recent improvements in the I/O patterns of experiment software. Results are presented for both live production systems and technologies not currently in widespread use. Performance is studied using tests, including real LHC data analysis, which can be used to aid sites in deploying or optimising their storage configuration.
[1] F. Rademakers,et al. ROOT — An object oriented data analysis framework , 1997 .
[2] Mike Kenyon,et al. Establishing Applicability of SSDs to LHC Tier-2 Hardware Configuration , 2011, ArXiv.
[3] Ilija Vukotic. Optimization and performance measurements of ROOT-based data formats in the ATLAS experiment , 2011 .
[4] Johannes Elmsheuser,et al. Functional and large-scale testing of the ATLAS distributed analysis facilities with Ganga , 2010 .