Dear RunAI users,

The scheduled maintenance on the RunAI cluster has been successfully completed. Below is a summary of the updates and key information:

What was done:
• Upgraded RunAI from version 2.20 to 2.24
• Upgraded Kubernetes from version 1.33 to 1.35
• Updated core Kubernetes components, including nginx-ingress, Prometheus, and the GPU operator
• Standardized the workingDir to /myhome for all job types
• If you need to modify it, you can use the CLI option: --working-dir /<NEW_PATH>

What’s new:
• Access to a Bash terminal directly from the Web UI
• Ability to submit jobs via the Web UI using YAML
• Full list of updates available in the official RunAI documentation: https://run-ai-docs.nvidia.com/saas/getting-started/whats-new-for-nvidia-run-ai-saas

Thank you for your understanding and patience as we continue to improve the platform.

Dear RunAi users,

We would like to inform you that a scheduled maintenance will take place starting:

• Monday, March 30 2026 from 09:00 that will take until Tuesday March 31 at 18:00.

⚠️ Please note that all workloads must be preempted before the upgrade.

We kindly ask you to stop your workloads in advance to avoid any disruption.

Thank you for your understanding and cooperation.
SDSC Helpdesk