[Hpc-notice] Slow login nodes during the weekend
Paul Van Der Mark
pvandermark at fsu.edu
Mon Apr 15 11:53:49 EDT 2019
Dear RCC users,
During this weekend we found that our login servers had become rather
unresponsive. This was caused by a large number of user processes. We
understand that it sometimes might be easier to run a small test on the
login nodes, but keep in mind that these are shared resources and
anything you do will impact the work of other users.
- Please use our cluster for jobs that take more than 4 cores or run
longer than an hour.
- Running an rsync for data movement can be taxing on our login nodes
too, we suggest you userun rsync to our export.rcc.fsu.edu node and/or
use globus.
- We are working on our job submission script generator, so you can
submit jobs directly from our website. In the mean time, please have a
look at https://rcc.fsu.edu/submit-script-generator
Don't hesitate to ask for help with your slurm job submission scripts,
we are here to help you.
Thank you,
Paul
More information about the Hpc-notice
mailing list