To all HPC users,

Based on guidance provided by the Shared Infrastructure Governance subcommittee on advanced GPU resources, the DoIT Research Computing Team has made some changes to the Slurm model that governs chip-gpu. In this new Slurm model, all GPU machines are available for general use so that resources never sit idle, but machines contributed by funds from specific research groups can opt for dedicated access to those machines through Slurm. Jobs running on contributed machines may be preempted if and when a user from a contributing research group opts for this dedicated access. For jobs run by all other users, these jobs may occasionally be preempted if the job is running on a contributed machine.

In this way, all users have access to all GPU machines, but users belonging to contributing research groups retain dedicated access to the contributed machines.

Users will notice that the chip-gpu Slurm cluster now shows three partitions. The default gpu partition encompasses all GPU machines, the gpu-contrib partition encompasses only machines where jobs may be preempted, and the gpu-general partition encompasses all machines where jobs will not be preempted due to dedicated contributor access. Note that the gpu-contrib partition is meant only to show users which nodes may see jobs preempted due to dedicated contributor access.

The three day preemption rule is unchanged across chip-gpu.

For more information about any of these hardware partitions and how preemption works across the chip HPC cluster, please see the wiki page: https://umbc.atlassian.net/wiki/spaces/faq/pages/1249509377/chip+Partitions+and+Usage

As always, let us know if you notice any issues or otherwise have questions or concerns by submitting a descriptive help request form found at the following link.

https://rtforms.umbc.edu/rt_authenticated/doit/DoIT-support.php?auto=Research%20Computing

Thanks for reading,

Roy Prouty

Assistant Director for Research Computing

(& Max!)

Hi everyone,

I’m writing to give everyone an update regarding the UMBC HPCF, particularly the new Retriever Research Storage System (RRStor). The Research Computing team within DoIT works to send these newsletters to keep everyone using our infrastructure informed and aware of any changes and new support structures we generate. 

System Status

The chip HPC is operating normally and is fully operational.

Updates: Migration of Storage Volumes

University budget cuts have compelled us to carefully examine the enterprise storage systems we maintain. While the new Ceph Storage Cluster (RRStor)  is here to stay, the Isilon storage system will be sunset at the end of this calendar year. The Isilon storage system was purchased at the start of the COVID-19 Pandemic, but its proprietary file system, high cost of expansion, and high software support costs make it no longer feasible for us to maintain.

Instead, we’ll be taking a portion of the saved expense from the Isilon and dedicating it to expanding RRStor and its capabilities. This will require that we migrate research storage volumes from the Isilon Storage System to Ceph before the end of the calendar year. This represents the careful movement of more than 2PiB of data, so DoIT Research Computing staff will be working with researchers to schedule the migration of their storage volumes. Note that the duration of each volume migration will depend on the size of the volume.

Research Volume Name:To avoid conflicts with volume names as this process begins and to avoid future confusion with group names and research volume names, we’ll be deploying new research volumes with the following syntax: “/umbc/rs/groupName” . So group “pi_doit” will find its data under “/umbc/rs/pi_doit” and group “nsf2346667” under “/umbc/rs/nsf2346667” . Each faculty PI group storage quota will be at least 10TiB.

Migration Process:Our plan for migrating data from the old system to the new system would entail the Research Computing team contacting each PI and scheduling a transition date before October 15. Before this date, the Research Computing team will make a copy of the research volume on the new system. On this date, the Research Computing team will disable the old volumes and the new volumes will be active. A month later, the old volumes will be deleted as we disable the entire old storage system. In most cases, there should be nothing for individual users or PIs to do when migrating this data.

User Support: Introducing Drop-in Office Hours

We’re pleased to announce a series of drop-in office hours that will occur each week on Mondays in ENGR 102 from 1500ET-1600ET. Participants may join in person or virtually. See the HPCF myUMBC Group Events page for more details. Otherwise …

• Please continue to make use of the HPCF Office Hours.

• We’re always working to make our Wiki Documentation better for users. Please let us know if we’ve missed something.

• Stay tuned for more in-person training and tutorials.

SIG

The SIG-CPU and SIG-GPU Committees have met and selected members. Please see the Shared Infrastructure Governance webpage for more information. The Slurm workload manager and faculty contribution models are among the first topics discussed in these groups.

Publications

If you have any publications, presentations, theses, or other works that made use of the campus cluster(s), please submit an RT Ticket with bibliographic information so that we can accurately reflect this work in our records and on the HPCF Website.

Need Help?

As always, please communicate any issues/questions to the Research Computing RT Queue (hpcf.umbc.edu > User Support > Request Help). 

Hi everyone,

I’m writing to give everyone an update regarding the UMBC HPCF, particularly the new Retriever Research Storage System (RRStor).

System Status

The chip HPC is operating normally and is fully operational.

Updates: 10TiB Storage Allocations for All HPCF PIs

We’ve expanded our enterprise research storage infrastructure with funds from last year’s NSF CC* Award! All research groups now have a 10TiB default quota. The DoIT Research Computing Team will be working with each research group to migrate to the larger volumes. Please note that this change to the default quota is for research storage directories shared by entire research groups and not for individual users’ home directories. Every user home directory has a default quota of 500MiB.

Pro Tip: Any user can use the “hpc_tools/quotas” module to see how much of their storage quotas have been used. This runs by default when a users logs in.

User Support

• Please continue to make use of the HPCF Office Hours

• We’re always working to make our Wiki Documentation better for users, please let us know if we’ve missed something

• Stay tuned for more in-person training and tutorials

SIG

The first meetings of the SIG-CPU and SIG-GPU subcommittees have met. For members of the SIG-GPU subcommittee, please be sure to review the most recent meeting notes sent by Karuna.

Publications

If you have any publications, presentations, theses, or other works that made use of the campus cluster(s), please submit an RT Ticket with bibliographic information so that we can accurately reflect this work in our records and on the HPCF Website.

Need Help?

As always, please communicate any issues/questions to the Research Computing RT Queue (hpcf.umbc.edu > User Support > Request Help). 

Thanks for reading,
Roy Prouty
Assistant Director for Research Computing
UMBC DoIT