I hope everyone has a good weekend.

At 16:55, large portions of the taki cluster were abruptly taken offline by administrators to help with an unexpected chiller failure in the data center. 

This unfortunately means that all running jobs were lost. 

I will send another message when the issue has been resolved.

Happy Friday!

Roy Prouty

UMBC HPCF System Administrator

To all taki users,

The maintenance downtime for the taki cluster environment is complete. Feel free to begin using the system.

We are aware of a few nodes that are down in the batch partition. These nodes are down due to a network hardware issue that will be resolved soon.

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

Roy Prouty

HPCF System Administrator

To all taki users,

On Wednesday May 31, 2023 taki will be down and inaccessible from 3:30pm - 5:30pm EDT.

Once our work is done on Wednesday, we will make another announcement.

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

Roy Prouty

HPCF System Administrator

To all taki users,

At 12:10PM taki-usr1.rs.umbc.edu of the taki cluster environment became unavailable for a few minutes. This login node was available again at 12:16PM. Any running sessions on this node at the time of the reboot were lost. No scheduled jobs were affected.

We are working to determine the cause of this unexpected downtime and apologize for any interruption to researchers' workflows.

While we don't expect this behavior in the future, remember that taki-usr2.rs.umbc.edu is also available as a login node for the taki cluster environment.

As always, it is important that all users run compute-intensive jobs through the SLURM scheduler so as to not create unnecessary load on the login nodes. Further, if you have experienced any other issues during this time--or have any other questions, comments, or concerns--please submit a descriptive help request via the form found at the following link.

Roy Prouty

HPC System Administrator

Dear HPCF Community,

Right now, Tuesday, 09/27/2022, 12:00noon-01:00pm, virtual online meeting, we have the chance to talk Intel engineers about their products.

They will provide an update on HPC technologies across Xeon CPUs, HPC software and compilers, AI optimizations, future roadmaps and more.   Please come with any questions, they want this as interactive as possible.

If you receive this as an e-mail, go to myUMBC -> Groups -> HPCF to find it as a post. The following link should get you the Webex meeting:

E-mail me with any questions at gobbert@umbc.edu

And by the way, always remember to send me information on publications (including senior, MS, and PhD theses)!  Documenting them on the webpage is a key measure of our success.

Matthias Gobbert, Department of Mathematics and Statistics

Chair Governance Committee

Dear HPCF Community,

This upcoming Tuesday, 09/27/2022, 12:00noon-01:00pm, virtual online meeting, we have the chance to talk Intel engineers about their products.

They will provide an update on HPC technologies across Xeon CPUs, HPC software and compilers, AI optimizations, future roadmaps and more.   Please come with any questions, they want this as interactive as possible.

If you receive this as an e-mail, go to myUMBC -> Groups -> HPCF to find it as a post. The following link should get you the Webex meeting:

E-mail me with any questions at gobbert@umbc.edu

And by the way, always remember to send me information on publications (including senior, MS, and PhD theses)!  Documenting them on the webpage is a key measure of our success.

Matthias Gobbert, Department of Mathematics and Statistics

Chair Governance Committee

To all HPCF users,

Our work bringing the HPCF research storage servers up-to-date has concluded. Please begin normal use of the clusters or other research computing machines.

Please let us know if you have any questions or concerns by submitting a descriptive help request form found at the following link.

Roy Prouty

HPC System Administrator

To all HPCF users,

On Tuesday September 13, 2022 both HPCF clusters (ada & taki) as well as the other compute systems that make use of HPCF Research Storage will be down and inaccessible from 3pm - 5pm ET. 

Once our work is done on Tuesday, we will make another announcement.

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

Roy Prouty

HPCF System Administrator