Research Computing at West Virginia University¶

West Virginia University Research Computing (WVU-RC) is a team inside WVU’s Research Office dedicated to supporting, enabling and advancing computational research at WVU.
WVU Research Computing maintains a portafolio of infraestructures to support its mission. We maintain several High Performance Computing (HPC) Clusters from general purpose to specialized ones, both on premises as on the cloud. WVU Research Computing also provides other services such as a large research storage facility called DataDepot, and a Demilitarized Zone (DMZ) for high speed data transfers called WVU Research Exchange (REX).
In addition to maintaining these facilities, WVU Research Computing offers support, consulting and training in areas of High Performance Computing, Data Analysis, Machine Learning and Parallel Programming.
The table below shows our portafolio of HPC resources (past and present):
Infraestructure/
HPC Clusters
|
Compute
Nodes
(total/active)
|
CPU
Cores
(total/active)
|
Accelerators
GPUs
(total/active)
|
Description
|
---|---|---|---|---|
Mountaineer
(decommissioned)
|
32/0 |
384/0 |
0/0 |
First centrally managed HPC Cluster for WVU.
CPUs from Intel Westmere microarchitecture (32 nm).
Decomissioned on 2018.
|
Spruce Knob
(end-of-life
decom. 2023)
|
176/148 |
3376/3036 |
14/5 NVIDIA GPUs
Tesla K20m
Tesla K20Xm
|
General-purpose HPC cluster first commissioned in 2017.
Heterogeneous cluster with Intel processors.
Sandy Bridge, Ivy Bridge, Haswell and Broadwell.
|
Thorny Flat
(in production)
|
178/178 |
6516/6516 |
47 NVIDIA GPUs
P6000 (21)
RTX 6000 (24)
A100 (2)
|
General-purpose HPC cluster.
Intel processors with Skylake and Cascade Lake.
Installed at Pittsburgh Supercomputer Center.
|
GoFirst
(in production)
|
Virtual Infraestructure running on AWS.
Serves Business Data Analytics (BUDA) program.
Chambers College of Business and Economics.
|
|||
Dolly Sods
(to be installed
August 2023)
|
37 |
1248 |
155 NVIDIA GPUs
A30 (120)
A40 (19)
A100 (16)
|
GPU Accelerated HPC Cluster.
To be provisioned in August 2023.
|
WVCTSI Cluster
Secure Cluster
(in production)
|
8/8 |
320/320 |
4 NVIDIA GPUs
Tesla V100S
|
HPC Cluster for use with
Protected Health Information (PHI)
HIPAA compliant
|
The contents of this website can be downloaded as a single PDF here: docs_hpc_wvu.pdf
There are several websites associated with WVU-RC activities, here is a list of the most relevant ones:
- The official webpage in Research Office portal about the Research Computing Division
- The legacy documentation was a Wiki website that will continue to be online for a while
- The HelpDesk ticket system
If your research was possible thanks to the use of our clusters please acknowledge the support using these comments: For Spruce Knob:
“Computational resources were provided by the WVU Research Computing Spruce Knob HPC cluster, which is funded in part by NSF EPS-1003907.”
For Thorny Flat:
“Computational resources were provided by the WVU Research Computing Thorny Flat HPC cluster, which is funded in part by NSF OAC-1726534.”
For Dolly Sods:
“Computational resources were provided by the WVU Research Computing Dolly Sods HPC cluster, which is funded in part by NSF OAC-2117575.”
For requesting help, create a new ticket at the Research Computing HPC Help Desk web page. You are welcome to e-mail any member of the WVU-RC team directly, but since we are not always at our desk, the ticket system will guarantee that your support question will be seen by someone currently available.
- Main Responsible for Documentation and Scientific Outreach
- Introduction
- Quick Start
- Basic Usage
- Advanced Usage
- Scientific Programming
- Software Administration
- Editing these documents
- Installing Packages in User Locations
- Linear Algebra
- Boost 1.79
- Message Passing Interface
- HDF5 and NetCDF
- Fast Fourier Transforms
- Force Field Molecular Dynamics
- CHARM++ and NAMD
- Density Functional Theory
- Big Data
- Python 3.9.7
- Matlab N-D multithreaded matrix operations (MMX)
- Building Julia
- Tinker9
- Updating NVIDIA Driver and CUDA Toolkit
- Domain Specific Details
- Clusters Specifications
- References