Available Support¶
Find out how to use our support portal effectively.
Join the CoSTAR compute users community on Microsoft Teams.
Information on Research Software Engineering support.
New to Slurm and HPC workflows? We prepared some examples to get you started on setting up your compute jobs.
Peer Support¶
The CoSTAR National Lab Compute Teams channel has other active users and updates from the Research Computing team about upcoming maintenance and other important information. Interacting with other CoSTAR Cluster users on Microsoft Teams can be a great way to get support and share insights.
Get involved by clicking this link to join the Teams Community
Research Software Engineering Support¶
If you require support from a CoSTAR affiliated Research Software Engineer for your project, e.g., to troubleshoot, you can reach out via Teams Community or the Ticket System
Support Portal¶
Use the support portal to send support requests to the CoSTAR support team.
When seeking out support please bear in mind that the Research Computing and Research Software teams operate from 9am to 5pm Monday through Friday, except for holidays.
Requesting New Software¶
The general policy on the CoSTAR cluster is for users to install and manage their own software where possible. Because every research project has unique dependencies, centralised system-wide installations are reserved for core infrastructure or broadly shared GUI applications.
Before contacting the Research Computing or CoSTAR RSE teams about new software, please work through this checklist:
Use containers (recommended)
Package the required software inside a container and run it on the cluster. This approach gives you full control over the software environment and makes your workflow portable and reproducible across different systems.
Install software in your home directory
Install user-level software directly in your home directory, for example by installing Miniconda to manage your Python environments using Conda. This approach is suitable for lightweight workflows.
Request installation on Open OnDemand (OOD)
If you require shared interactive applications (e.g. GUI-based tools such as VS Code or MATLAB) to be made available through Open OnDemand, in addition to the existing ones, you can request this by raising a support ticket. The Research Computing team will assess feasibility and suitability for cluster-wide use.
Still stuck?
If your software has complex licensing requirements, requires deep system integration, or if you have genuinely hit a wall trying to containerise it, please contact the CoSTAR RSE team or open a support ticket.
Note: When submitting your ticket, please mention which self-service methods (Containers/Conda) you have already tried so our team can assist you more efficiently.