Computational Onboarding for SDSS Postdocs: An intro to using the serc partition

Welcome to the SDSS Center for Computation! We provide many computing resources including access to the serc partition on Sherlock, a supercomputer here on campus. Using serc give you priority access to compute machines within Sherlock. Today we'll get you started on using serc resources. You can use serc in an interactive capacity, which allows you to interact with the Sherlock supercomputer and the Oak storage space as if you were using your own laptop, or you can schedule a job using Slurm, which provides with larger and longer access times.

Live Code Agenda:

1. Open an interactive session, pointing to Oak
2. Create a new notebook, and use Globus to confirm that the notebook file is saved in Oak
3. Clone this GitHub repo into Oak and interact using serc
4. Run an example notebook, and save the outcomes to Oak
5. Write a script file based on analysis pipeline in the notebook
6. Write an sbatch file for your job, and schedule the job using Slurm

Do you have any questions?

Please reach out to us at sdss-compute@stanford.edu, or schedule a consultation with our team here.

Would you like to provide feedback on your SDSS-CC onboarding session?

Please provide anonymous feedback on your onboarding here.