Here is an outline on how to get started on Falcon

  1. Acquaint yourself with policies and documentation regarding the cluster
  2. Request access
  3. Connect to the cluster
  4. Submit jobs

Policies and Conduct

On this page, we outline specific requirements from your side and best practices that are essential for maintaining a productive shared working experience. These policies are in addition to the CS computing policies described here.

General

  • A basic knowledge of Linux command line and shell scripting is expected. Consider going through some Linux tutorials first if you have never used the command line.
  • Make use of the documentation on the SNA website before seeking assistance. Everything required to write and submit jobs for many applications is described here.

Jobs

  • Submit jobs to the appropriate partition, and with the correct QOS.
  • Please make sure correctness of your code/algorithm by testing on a smaller dataset/time frame.
  • Verify that your job completes successfully before submitting it several times.
  • Make sure paths referenced in the job script are valid.
  • Using checkpointing mechanisms is highly recommended.

Resource Usage

  • Jobs should not be run directly on the login nodes, also known as submit or user nodes.
  • Only request the resources which your job needs, do not over-request resources.

Requesting Access

Access to the Falcon cluster is open to all CS faculty, and to any CS student with a CS faculty endorsement. Your CS account needs to be enabled for cluster access before you can start using Falcon.

Send your request via email to compsci_sna@cs.colostate.edu and include:

  • A brief explanation of how you would use the cluster for your research.
  • An endorsement from your research advisor.

Connecting to Falcon

To connect to the Falcon cluster use an SSH client from your machine and connect to the host falcon.cs.colostate.edu, using your CS Linux account credentials.

You may use any of the following SSH clients:

  • ssh from a Unix/Linux computer
  • ssh from a terminal window on a Mac computer
  • PuTTy from a Windows computer
  • ssh from a Windows PowerShell on a Windows computer

If you need help with SSH, refer to our SSH guide here ../remote-connection/ssh

Submitting Jobs

To submit jobs on a cluster, one needs to allocate resources and then run the executable over these resources. This is usually done by writing a job script. In this job script, you specify a partition, the resources your job needs (CPUs, memory, GPUs, etc) and a QoS.

Refer to the Sample Job Scripts section for help on writing job scripts.

Submit your Job

After you have your job script written in a file (say job.sh), submit the job via the sbatch command as shown below:

sbatch job.sh

On successful submission, you would see an output similar to

Submitted batch job 1025

Where that number 1025 is the job id number. This is job id number is needed for troubleshooting, so make a note of it.

You can then monitor the status of your job, refer to the Monitoring Jobs section.

Click/Tap here to return to Falcon HPC Cluster page.