Computing Cluster

High-Performance Computing (Sun and HP Hardware)

Our high-performance computing (HPC) is used for developing and scaling up technologies for energy and environmental processes using computational fluid dynamics modeling. Please click here for usage rates.

We operate two types of HPCs in our facility. The first type of HPC is a “Beowulf” style cluster that is Linux based. The other type is a a Windows terminal server (2008) – high-performance computer server (HPCS). By offering these two types of HPC, we can deal with the vast majority of requests from our user base.

The hardware for our Linux-based system is composed of Sun (V20Z) and HP (Opteron). We currently have 18 compute nodes (4 cores at 2.2 GHz with 8 GB memory). Employing our job scheduler, we try to optimize the use of these HPC resources. We are interested in offering the quickest possible turnaround times for our users. OSCIEF supports two type of jobs: parametric sweeps and symmetric multiprocessing (SMP). Parametric sweeps allow users to run the same algorithm (parallel jobs) while varying a large number of variables or parameters. Using SMP, a user can simultaneously run one large job accross serveral processors. SMP can be combined with the parametric-sweep method. Please contact the OSCIEF HPC specialist to discuss your specific application and what we can offer you.

The hardware for our HPCS is a Windows Server 2008 Terminal Server. This hardware allows Windows users to access our HPC within a familiar graphic user interface. We have 16 cores (72 GB RAM) to process your job.

Available Software

Sun Grid Engine Job Scheduler
Gromacs
NAMD
VMD
OpenMPI and MPICH MPI Libraries
Fortran, C++, and C Compilers
COMSOL
CanvasX
Autodesk
Scripting Templates and Debugging Services Available
If you require software not on this list, please contact us to discuss your software needs.