The HENP computing facility at Brookhaven National Laboratory supports both the Relativistic Heavy Ion Collider (RHIC) and US involvement in the ATLAS LHC experiment. The facility includes 150 TBytes of centralized online (disk) storage, which is served to a processor farm of 2000 CPU's. Multiple levels of virtualization are used in the deployment of this storage. This allows for a great deal of flexibility in the maintenance, performance tuning, and expansion of the resource in a manner which is transparent to it's users. The tools and strategies employed and the additional functionality achieved will be discussed.
Cite
@article{arxiv.physics/0306026,
title = {Large Scale Online Storage Management},
author = {Maurice Askinazi and David Free and Bruce Gibbard and Thomas Throwe},
journal= {arXiv preprint arXiv:physics/0306026},
year = {2007}
}
Comments
Talk from the 2003 Computing in High Energy and Nuclear Physics (CHEP03), La Jolla, Ca, USA, March 2003, 6 pages