| HPC at UNBC |
FAQS
|
|
Frequently Asked QuestionsWho is eligible to have an HPC account? How may I apply for an HPC account? How may I stay updated about the HPC systems? How may I get training to use the HPC systems? How may I know if the compute cluster is up and functioning normal? Some useful tips while working with research storage shares (ceph). Who is eligible to have an HPC account? If you are a graduate student, post-doc, staff or faculty member at UNBC, you are eligible to have an HPC account. Undergraduate students who are doing a research project supervised by a UNBC faculty member are eligible for an HPC account. How may I apply for an HPC account? Request Service through this link. How may I stay updated about the HPC systems? Subscribe to the hpc-users list by going to http://lists.unbc.ca/mailman/listinfo/hpc-users. You will receive email updates on important HPC related news like server maintenance, new software licenses, storage updates, data migration, workshops, etc. How may I get training to use the HPC systems? Contact nikhil.aravindakshan@unbc.ca for basic instructions and guidance to get started with the HPC systems. He is also available to answer any questions you might have about HPC usage at UNBC. There are also workshops happening every year, which are notified through the hpc-users mailing list (subscribe at http://lists.unbc.ca/mailman/listinfo/hpc-users) and the HPC webpage (https://web.unbc.ca/~hpcweb/hpc-workshops.html), which also has recordings of past sessions. How may I know if the compute cluster is up and functioning normal? There are some common commands you can use at the command line to check the status of the cluster once logged into it: uptime tells you how long the cluster has been up since the last downtime or reboot, df -h is useful to see if all the mount points (like storage drives) are mounted, top gives you a list of processes using various amounts of CPU and memory, which is useful to know when you experience a system slowdown (press q to quit once done), sinfo gives you a brief report on the usage of the nodes for various jobs by different users, squeue shows you the status of jobs in the queue from various users. Some useful tips while working with research storage shares (ceph). See the Knowledge Base article in this link.
|