Dec 15, 2025

On Saturday, 12/06/2025, a brief outage occurred on the file system hosting research spaces. This outage reoccurred on Monday, 12/08/2025. To mitigate impacts to our users and their running jobs, we disabled some features on the file system at that time. Beginning Thursday, 12/11/2025, we began reenabling all file system features. Due to the work on Thursday, 12/11, periodic slowdowns and pauses on the file system have been reported. We are aware of this issue, and are actively working with our vendor to resolve these slowdowns and pauses as quickly as possible.

Jan 13, 2026

Slurm job IDs have reset back to 1 after surpassing the maxumum job ID of 67,043,328. This has occurred every few years and is expected behavior that should not have any adverse impact on jobs. If you have any quesions about this reset, please contact us.

Jan 5, 2026

Happy New Year! Per ICER policy , users who are not associated with a buy-in account are limited to 500,000 CPU and 10,000 GPU hours per year. With the new year comes new beginnings as these CPU/GPU hour limits are reset. If you have any questions about this reset, please contact us.

Nov 10, 2025

We have have updated scheduler parameters that determine how users’ recent cluster usage affects their FairShare priority factor. Previously, memory usage was weighted higher than CPU usage and GPU usage was not considered. Now, CPU, memory, and GPU usage are all considered in the calculation of FairShare priority and weighted in proportion to the overall availability of those resources.

More information about this change can be found in our documentation

If you have any questions about this change, please contact us