S13: HPC Archive Solutions Made Simple
Sunday, Nov. 13
1:30-5:00
Alan Powers, CSC
Jason Hick, NERSC
Matt Cary, NASA Advanced Simulation Facility
http://sc11.supercomputing.org/schedule/event_detail.php?evid=tut168
ABSTRACT:
This tutorial provides an overview of the different hardware technologies and software features used in HPC archival storage systems. The tutorial will touch on the following topics: tape libraries, tape drives, disk cache, HSM, data movers, migration, data recall, storage policies, tape pools, tape compaction, tape striping, archive metrics, data integrity, importing and exporting data, and disaster recovery. Other topics of interest developing an RFP. # How to determine recall and migration policies by profiling existing data files? # What requirements to consider for architecting an archive system? # What future requirements should be considered when sizing an archive system? Highlights of the archive policies of four major HPC sites: National Energy Research Scientific Computing Center (DOE/NERSC - 22 PB), NASA Advanced Supercomputing Facility (NASA/NAS - 28 PB), NOAA Environmental Security Computing Center (NOAA/NESCC - 14 PB), and Geophysical Fluid Dynamics Laboratory (NOAA/GFDL - 28 PB) - will be presented.


