High-Performance Computer Cluster Overview

Topology

The CoE-HPC cluster currently has:

Head node with 12 cores
8 servers with 16 cores each
4 servers with 20 cores each.
208 total computational cores in the cluster
Total RAM of 8x128GB or 1024GB (not including the head node)

Allocations

Each user receives a default allocation with the following directories:

$HOME directory: the home directory can be used to host critical files like software applications. It is backed up to a remote location. HOME directories are private to each user.
data: the ‘“data” file system is backed up to a remote location.
work: a group directory where members of a research group can easily share files. Not backed up.
scratch: a local directory for each user. All data on this directory is private to the user. Not backed up.

Job Submission

The CoE-HPC uses SLURM (Simple Linux Universal Resource Manager) to manage resource scheduling and job submission. SLURM uses “partitions” to divide types of jobs.

Current partitions at the CoE-HPC: ($> sinfo)
Partition	Time limit (hours)	Number of nodes	Default/Max Cores per Node	Max memory per node (GB)
Debug	2:00:00	001-004	1/16	128
defq*	unlimited	001-004	1/16	128
Shared	unlimited	005-008	1/20	128
Fast	unlimited	009-012	1/16	96

Software

LAMMPS
Quantum Espresso
VASP
MATLAB
Akantu
Variety of in-house codes
ANSYS (available soon)
Abaqus (available soon)TensorFlow (available soon)

Acknowledgement in Publications

All users of the computational resources of the CoE-HPC cluster for their research will provide an acknowledgement statement in their publications. Any of the following three options, or variations therein, are allowed:

Option 1: “This work used the High-Performance Computing cluster of the College of Engineering at Villanova University.”
Option 2: “Computer time allocations at Villanova’s HPC-CoE cluster are acknowledged.”
Option 3: “All simulations were run on the High-Performance Computing cluster of the College of Engineering at Villanova University.”