Cluster usage

From systems
Jump to: navigation, search

Policies & Procedures

Proceedures

  • All users who wish to use cluster computing resources at ARCS must request access through their respective Investigator or Department Administrator. E-mail requests to services [at] c2b2.columbia.edu.
  • All cluster computing usage is billed on a quarterly basis to the department or lab based on a negotiated rate. First time users may be required to purchase a startup package of CPU time and may be subject to account creation fees.
  • Cluster accounts shall be created once payment for usage time is paid in full.
  • Technical documentation can be found on the ARCS wiki page named "Titan Cluster" in the Documentation section. http://wiki.c2b2.columbia.edu/systems/index.php/FAQ/Titan_cluster .
  • All cluster users must adhere to the cluster usage guidelines stated stated below.

Suggested practices

The recommend that you follow the following common sense practices when using the cluster computing resources.

When you run jobs, check your jobs regularly to see:

  • How much memory your jobs are using.
  • How much time they are taking to run.
  • How much disk space are they using (for writing files).
  • How many processors are you using at one time.
  • How many jobs you have waiting in the queue.

Rules for use

The following practices are mandatory for all users of the cluster computing resources.

  • Do not store files on the cluster node disks. Clean up temporary files written there. These disks are temporal and get overwritten when a cluster node is rebooted.
  • Do not run multiple computationally intensive tasks at the same time under a single job (unless you are in a parallel environment, e.g. mpich or smp).
  • Do not run jobs on the login nodes. These nodes are only for submitting jobs. Test and run jobs on the cluster nodes. When you submit a job to the cluster, you should check to see that it is running on the cluster nodes.
  • Do not attempt to circumvent security, restrictions or resource allotments.
  • This is a shared use system. Everyone must be respectful of other users and their right to use the system. Any attempt to "hog" resources or prevent other users from fair use of the system will not be tolerated.

Usage violations

Users failing to follow the cluster guidelines will be subject to a verbal and/or written warning. Continued abuse of the cluster will result in account deactivation and a written report to the lab's PI. Reinstatement of the cluster account will be granted once the offending user and their PI acknowledge the violation and agree to adhere to the cluster usage guidelines in writing.

Further violations regarding cluster misuse will result in account deletion, data confiscation and the reimbursement (if any) of the remaining cluster time purchase by the lab.