Resources
My research group uses a Linux-based High Performance Computing Cluster (HPCC) to model atmosphere and biosphere processes and to process and analyze modeled and measured atmospheric data. The cluster is made up of 14 nodes, each consisting of Intel-based 3 GHz quad-core Xeon processors containing a total of 112 processors. Each node has 16 GB of RAM and the cluster has 15 TB of storage using high-speed SAS drives. A pending expanision in 2008-2009 will add an additional 16+ nodes and ~10-15 TB of storage space.
The cluster is maintained by the University of Minnesota's Office of Information Technology and is housed in the West Bank Office Building (WBOB) on the West Bank campus in Minneapolis.
The cluster runs the Redhat Linux operating system, version 4.5 and code runs with both OpenMP and MPI library specifications. We use MAUI and TORQUE to handle job submission and scheduling. A wide variety of applications are run on the cluster including:
- The Community Climate System Model (CCSM3)
- The Community Atmosphere Model (CAM3)
- The Community Land Model (CLM3)
- The Integrated BIosphere Simulator (IBIS)
- The Weather Research and Forecasting Model (WRF), version 2
- NCAR Command Language (NCL)
- Ncview
- netCDF Operators (NCO)
- R - Statistical computing package
In addition, there are several commercial applications that are used for programming, data processing, and visualization:
- Portland Group PGI Server Complete - Fortran 77 and 95, HPF, C, C++ compilers, and OpenMP debugger and profiler
- Interface Description Language (IDL) - for data processing and visualization
- Matlab - for data processing and visualization
We also have ample desktop and laptop computing (Mac or Windows) for cluster access, communication, writing, data processing and visualization, etc. And no, we won't don't use a TRS-80!