Compute
Overview
The Notre Dame Human Neuroimaging Center (ND-HNC) provides computational resources to support a wide spectrum of neuroimaging analyses. Our resources are structured into three levels to accommodate everything from initial data exploration and single-subject processing to intensive, large-scale group analyses.
| Resource Type | Location | Hardware | Primary Purpose |
|---|---|---|---|
| Local Workstations | ND-HNC B30 | Dell Pro Max Linux Towers, Mac Studios |
Sandbox testing, single-subject processing, pipeline development. |
| Dedicated Server | vfpc-mri-rh-l01.campus.nd.edu | Dell Rack (56-core Intel Xeon, 256GB RAM, 4TB Storage) | Resource-intensive local processing, multi-subject parallel tasks. |
| HPC Cluster | CRC | 1280 Priority-Access CPUs | Large-scale datasets, massive parallelization, complex modeling. |
Access
- Contact Nathan Muncy. HPC resources also require a CRC account.
Local Workstations
The ND-HNC provides a compute room (B30) populated with high-performance local workstations for MRI researchers. This helps avoid startup issues of hardware acquisition and software installations, and removes the overhead of cluster computing when processing data for only a few subjects. These workstations also support code development and sandbox testing of pipelines before their deployment on the cluster.
Dedicated Server
For resource-intensive processes that exceed standard desktop capabilities, the ND-HNC features a dedicated Dell rack server. This centralized, high-capacity machine is equipped with an Intel Xeon Gold 6330 CPU (56 cores @ 2.00GHz), 256GB of RAM, and 4TB of high-speed local storage. This is ideal for multi-subject parallel preprocessing or running intensive local pipelines that require substantial memory and multi-core processing.
HPC Cluster
To support large-scale studies, advanced machine learning applications, and complex network modeling the ND-HNC leverages the power of the university’s broader compute infrastructure. We have purchased 1,280 dedicated CPUs for the Notre Dame Center for Research Computing (CRC) cluster, granting members of our group priority access. These resources facilitate massive parallelization, enabling our community to scale their analyses for large datasets or complex worfklows.