The SoM Research Computing Center (SRCC) collaboratively works with the Duke research community to identify, design, deliver, and support technological solutions for biomedical research. The SRCC provides a full-service high performance computing (HPC) center, including one-on-one consultations, pipeline optimization, training, and data management strategies to researchers across all departments.
Services currently offered include both on-premises and cloud-based (Azure) options. Please reference the HPC Cost Model and Rates section for more details.
The on-premises HPC cluster operates in Duke Health data centers and offers both CPU and GPU-based computing options (see below). Additional hardware may be purchased by PIs to augment the cluster.
Storage for the on-premises HPC cluster consists of Home Directories and ~480TB of available Scratch Space storage.
- Home Directories:
- 50 GB quota per user
- Use for storing scripts, creating conda environments
- Scratch Space:
- 10 TB* quota per account
- No backups
- Files not accessed for 60 days are auto-deleted.
- Reminder email sent to account owner/users one week (7-days) prior to deletion.
PIs are responsible for ensuring that adequate supplemental storage which is external to the cluster is in place for their data.
*To request a one-time quota increase, please submit a request to: Request for High Performance Computing (HPC) Services | Support@Duke (service-now.com)
COMING SOON: list of recommended options and pricing for purchase of additional compute and storage hardware.
| Infrastructure | Details |
|---|---|
| CPU Compute Nodes | 27 Nodes (each with 128 threads, 1TB RAM) |
| GPU Compute Nodes | 1 Nodes (64 threads, 1TB RAM, 4 x NVidia L40S) |
| Storage | 560TB RAW Storage |
| Software | SchedMD Slurm Workload Manager |
| Networking | Infiniband Switch (200GB/s) |
Note: These are base specifications
The DHTS Azure HPC service is a platform for managing and operating high performance computing clusters. The design of this system allows for compute on sensitive and restricted data types, such as Protected Health Information. The DHTS Azure School of Medicine High Performance Computing cluster (DASH) is one cluster in the system, and is a shared resource for Duke faculty, staff, researchers, and students working with research related to the School of Medicine. A research group can also request a custom build-out of an Azure HPC cluster or other virtual machines for computational analysis with specific choices of VM types and scheduler software.