HCC will be retiring the Crane compute cluster on July 1st, 2023 and powering it off shortly after. This is being done to allow HCC to consolidate hardware into a single resource, improving performance and reliability.
Important Dates:
Reduced capacity beginning in May 2023
Crane Login Disabled on July 1st, 2023
You can move your data to $COMMON, Swan, or Attic. In any case, please make sure you have your important data backed up in another location. For large data transfers, it is strongly encouraged to use Globus to transfer your data. The Globus transfer servers for Crane, Swan, and Attic provide a faster connection for the data transfer and provide checks on both ends of the data transfer.
The retirement of Crane also provides a good opportunity for researchers to evaluate and archive data that is no longer being used.
Outside of the normal 6-month purge policy on $WORK, which will continue to be in place, Crane’s $HOME and $WORK will be removed when Crane is retired later in 2023.
We also have a brief overview of the use of Globus from a prior workshop available here.
Your workflows, including submit scripts and data, should be able to be directly copied over to Swan and run. Some items to double check would be partitions, job constraints, and specific software module versions.
There are no additional steps needed to access Swan with your HCC account. Swan can be logged in with SSH with swan.unl.edu and the OnDemand interface can be accessed at https://swan-ood.unl.edu.
We will no longer be accepting new partition leases for Crane, but will continue to do so for Swan. Any hardware on Crane that is still within the warranty period will be migrated to Swan to help consolidate resources into one pool, and continue to be available for use there.
We will be slowly migrating the good hardware from Crane to Swan over the next handful of months, which will resource the resource availability on Crane, while increasing the availability on Swan.