Tape Storage Optimization at BNL
David YU, Jérôme Lauret · Journal of Physics Conference Series · 2011
The BNL's RHIC and Atlas Computing Facility (RACF), is supporting the RHIC experiment as its Tier0 center and the Atlas/LHC as a Tier1 center. The RACF had to address the issue of efficient access to data stored to disk and tape storage. Randomly restoring files out of tapes destroys access performance to tape by causing too frequently, high latency and time consuming tape mount and dismount. BNL's mass storage system currently holds more than 16 PB of data on tapes, managed by HPSS. To restore files from HPSS, we make use of a scheduler software, called ERADAT. This scheduler system was originally based on a code from OakRidge National Lab, and then it was renamed to BNL Batch at 2005 after some major modifications and enhancements. The new BNL Batch, ERADAT, provides dynamic HPSS resource management, schedule jobs efficiently, enhanced visibility of real-time staging activities and advanced error handling, to maximize the tape staging performance. ERADAT is the interface between HPSS and other applications such as the DataCarousel, our home developed production system and dCache. Scalla/Xrootd MSS can also be interfaced with HPSS via DataCarousel. ERADAT has demonstrated great performance in BNL and other institute.