Rosen Center for Advanced Computing
Have a request for an upcoming news/science story? Submit a Request

Scholar Cluster Maintenance

  • Maintenance
  • Scholar

UPDATE:

The upgrade of the Scholar cluster has been completed, and job scheduling has been enabled for the gpu queue. Due to the diversity of hardware on Scholar, we will continue to perform additional application upgrades and testing in the coming days. The scholar CPU queue will become available following these. Major changes to the Scholar environment include:

  • Deployment of the Rocky9 operating system
  • Major upgrades to the GCC and Intel compilers and OpenMPI MPI libraries
  • Python and Anaconda can now be found under the conda module
  • The Intel compilers have migrated to OneAPI packages, which can be found under the intel module
  • The bioinfo module has been deprecated. Bioinformatics software maintained by RCAC can be found under the biocontainers module.
  • Home directory quotas are now enforced at 25GB, the same as on other RCAC clusters.

We are aware of some issues with the R and netcdf-c libraries on specific hardware types and are working on a fix. Users can run these applications directly in a Thinlinc desktop session in the meantime. If you notice other application issues, please report them to rcac-help@purdue.edu.

UPDATE:

We have made further significant progress on upgrading the Scholar cluster. Upgrade and testing of applications and services are mostly completed, but remain ongoing.

We will provide another update by 8pm.

UPDATE:

We have made further significant progress on upgrading the Scholar cluster. Operating system upgrades to the storage system have now been completed. Upgrade and testing of applications and services are ongoing.

We will provide another update by 5pm.

UPDATE:

We have made significant progress on upgrading the Scholar cluster. However, the upgrade process is taking longer than we anticipated. Re-racking and installation of new hardware has been completed, as have operating system upgrades on compute and front-end nodes. Operating system upgrades to the storage system are ongoing, as are upgrades to applications and services.

We will provide another update by 12pm tomorrow.

ORIGINAL:

The Scholar cluster is scheduled for maintenance and will be unavailable on Monday, December 16th, 2024 at 8:00am EST through Thursday, December 19th, 2024 at 5:00pm EST. During this maintenance, Scholar will receive hardware and software upgrades that will enhance its capacity and functionalities to better serve Purdue's instructional needs, Transformative Education 2.0, and the increasing number of courses using computational methods.

What is being upgraded?

During this maintenance, the capacity of the cluster will be increased through the addition of new nodes and filesystem performance upgrades. Scholar's operating system will be upgraded to Rocky Linux 9, which is effectively two versions ahead of the current OS, CentOS 7. System libraries will shift versions forward, and the applications, libraries, and compilers provided by RCAC through the "module" command will be upgraded. Other user experiences such as SSH, web access, and job scheduling will remain unchanged.

How does this affect you?

The Scholar cluster will be unavailable during the mainteance window. User files will persist through the upgrade, but Slurm jobs still queued when upgrades begin at Monday, December 16th, 2024 at 8:00am EST will not run. Node reservation will prevent submission of new jobs that are not expected to finish by the start of the maintenance.

After upgrades have been completed, it is recommended that users check needed applications and course materials for continued compatibility. It is also recommended that any Python, R, or other self-installed packages be reinstalled. Please reach out to rcac-help@purdue.edu if you have any questions about the Scholar upgrade or need support. Please note that RCAC staff will observe the university winter recess from December 23, 2024 through January 1, 2025.

Originally posted: