Welcome to the DKU Compute Cluster (DKUCC) user documentation site. You’ll find overviews, documentation, FAQs, and tips on the DKU Compute Cluster on these pages.
The DKU Compute Cluster is a general purpose high performance/high-throughput installation, and it is fitted with software used for a broad array of scientific projects. With a few notable exceptions, applications on the cluster are generally Free and Open Source Software.
Quick facts:
- 18 nodes which combined have 1672 CPU cores, 8*H20 141G, 16*A40 48GB, 40*L20 48G and 14.9TB of RAM
- Interconnects are 10 Gbps
- Utilizes a 400TB Isilon file system
- Runs AlmaLinux 9.3 and SLURM is the job scheduler
The DKUCC is managed and supported by the DKU IT.
Please note that due to limited resource, all data on DKUCC will not back up, please backup your data accordingly.
Important updates:
Please use those following latest hostname to access to DKUCC:
- For ssh to login node:dkucc-login-01.rc.duke.edu
- For OnDemand, please visit:https://dkucc-ondemand-01.rc.duke.edu
Important notice on ssh to login node
Linux and MacOS systems generally come with a SSH client already installed and you can ssh directly from a terminal window. For Windows users you will need an ssh client such as PuTTY. Advanced users may consider installing WSL (Windows Subsystem for Linux), see installation instructions here, but beware DKUCC support staff do not offer support for your local computer and software installations. Other ssh clients may work or not, but beware of clients that create persistent sessions to the DKUCC.
Users may connect to the DKUCC login nodes using VSCode to troubleshoot code and manage scripts. However, VSCode SHOULD NOT BE USED for computationally intense tasks on login node (dkucc-login-01.rc.duke.edu) or on compute nodes via common or scavenger partitions. If you would like to execute code using VSCode, connect to OnDemand and run your code interactively using the CodeServer option.
