Review Great Lakes HPC user guide for initial setup

[DaniJonesOcean]

Issue Description:

We're planning to use the University of Michigan's Great Lakes High-Performance Computing (HPC) system for this work. Start by reviewing the Great Lakes HPC user guide. This will ensure you understand how to utilize this service, and it should also get you started on account setup and maintenance.

NOTE: As you'll notice, U-M ITS considers the Great Lakes HPC platform to be a separate entity from the Turbo Storage data volume. In practice, some folks will refer to both as the "Great Lakes HPC".

Key Objectives:

• Review the user guide to understand the setup and usage process, as well as important parameters such as cluster defaults and partition limits.
• Familiarize yourself with the policies for using the Great Lakes HPC, including storage policies and job submission guidelines.
• Learn the processes for getting started via the web interface (Open OnDemand) and/or the command line (your choice, of course).
• Take note of any sections on GPU usage, job submission and management, data transfer methods, and interactive applications that might be relevant to our work.

After reviewing the guide:

• Ensure you have all necessary UMich accounts, Duo authentication, and the VPN needed for access.
• Bookmark the Great Lakes cluster cheat sheet for quick reference to commonly used Linux (Bash) and Slurm commands.
• Familiarize yourself with the guidelines for using interactive apps like Jupyter Notebook.

Deliverable:

• A brief checklist confirming that you've reviewed the guide and highlighting any questions or clarifications you may need. This can help us assist you in addressing any remaining uncertainties.

The user guide is a valuable resource that will support your success on the cluster. Make sure to check out the supplementary materials like the cheat sheet and the Open OnDemand guide linked within the user guide.

Resources:

• Great Lakes HPC User Guide
• Great Lakes Cluster Cheat Sheet (PDF)
• Slurm User Guide for Great Lakes
• Open OnDemand Guide

Video Training Resources: U-M has a number of training videos that cover various aspects of using U-M HPC. Consider watching some of them as you see fit.

[DaniJonesOcean]

@DaniJonesOcean Check the links in the above - they appear to be broken

[eredding02]

@DaniJonesOcean I have completed the majority of the getting started steps, such as getting Duo, Great Lakes login, VPN, and OnDemand access. With Dani Jones I was able to successfully run a Jupyter notebook, one of the interactive apps, using the HPC.

I am still having difficulty trying to submit a job via terminal and job composer. In the "Jobs" portion of OnDemand there is an sbatch error, which I also received during the Intro workshop when attempting to submit a job via terminal using nano.

[eredding02]

@DaniJonesOcean I successfully ran a job, a .sh file (echo "Hello World"), using both terminal and Job Composer.

[DaniJonesOcean]

@eredding02 Fantastic! Feel free to close this issue when you're ready.