Storing and Managing Scientific Data with HPSS

The High Performance Storage System has become the solution of choice for storing, archiving, and retrieving large quantities of scientific data. There are currently 41 instances of HPSS installed at 25 organizations around the world, totaling many petabytes of stored data and many millions of files.

Meet some of the HPSS customers who are utilizing the most powerful mass storage software in the industry today to manage their large scientific data archives.
The San Diego Supercomputer Center (SDSC) and Caltech store many hundreds of terabytes of scientific data in HPSS for a variety of government and university projects. One of those projects, Digital Sky, will store four million high-resolution digital images from the 2-Micron All Sky Survey (2MASS). SDSC manages one of the largest production installations of HPSS, currently storing about 300 terabytes.
The Lawrence Livermore (LLNL), Los Alamos, and Sandia laboratories of the DOE each use HPSS as part of the ASCI program to store hundreds of terabytes of simulation and other scientific data. The HPSS system at LLNL stores up to several terabytes of data per day.
Brookhaven National Laboratory (BNL) is using HPSS to store and manage large amounts of high energy physics data collected from four particle collision detectors that are part of the Relativistic Heavy Ion Collider (RHIC). BNL shares the data with the Institute of Physical and Chemical Research (RIKEN) in Japan by doing exchanges of data with the HPSS system at RIKEN. RHIC's 2.4 Mile Ring
NASA's Langley Research Center has built their large-scale Distributed Mass Storage System (DMSS) on HPSS. Some of the mission applications supported by the data stored in DMSS include Computational Fluid Dynamics solutions of Aircraft Operations, Propulsion-Airframe Integration, and Re-entry Vehicle Design. Re-entry Vehicle Design and Development at NASA's Langley Research Center
The French Commisariat à l'Energie Atomique - Division des Applications Militaires (CEA/DAM) will be using five StorageTek tape libraries with HPSS to store one petabyte of numerical simulation data per year. The CEA/DAM Computer Center is located at Bruyeres Le Châtel south of Paris.
Landfall of Hurricane Fran
NOAA's National Climatic Data Center (NCDC) is responsible for the nation's climate data. This includes the ingesting, processing, and archiving in HPSS of satellite, radar (NEXRAD), and global in-situ data from thousands of observing stations, platforms, and remote sensors. NCDC has created an on-line Web interface so users can order and retrieve data via the Internet.
Storing One Petabyte of Data per Year
For further information

Bob Coyne
(IBM) 281-335-4040
coyne@us.ibm.com

Harry Hulen
(IBM) 281-488-2473
hulen@us.ibm.com