10/16/2024
SCINet’s Ceres supercomputer has a new data storage system that offers substantial advantages over the old storage system. These advantages include nearly 75% more usable storage space (increased from ~5.5 PB to ~9.6 PB), an ~80% reduction in power and cooling requirements, and faster read/write performance. Ceres users will automatically benefit from these improvements because all file systems (/project, /90daydata, and /home) have been moved to the new storage system.
With the new storage system, please note that there is a change in how quota enforcement is managed which may affect some quotas. Under the previous system, quotas were enforced based on the physical space required to store data. For example, if 100GB of files only required 80GB of physical storage space, those files would only be counted as 80GB against your storage quota. With the new system, quotas are enforced based on raw file sizes. In the same example above, even if the storage system only needs 80GB of physical space to store 100GB of files, the files will still be counted as 100GB of data against the quota. If you find that you are now exceeding or approaching your /project quota, the Project PI or Manager can request a quota increase for your project.
Any compute jobs that were not able to complete by October 11 at 4:00 PM CT were held in the job queues after the maintenance completed. If you have a job that is being held, you will need to explicitly release it by running scontrol release <JobID>. If you wish to cancel the job instead, please run scancel <JobID>. (You can find the ID of your queued jobs by running squeue --me.)
If you believe you are missing any files, feel free to open a support ticket (email scinet_vrsc@usda.gov). Please note that the old /project filesystem will be available for a limited time from Ceres’ data transfer node (ceres-dtn) as /project-old. In the near future, this will be removed to free up space, power, and cooling for Ceres compute resources.