Quick Start¶
This is a short tutorial intended for first-time users. We assume no familiarity with High-Performance Computing (HPC). The only requirement is basic familiarity with your own computer in order to install the SSH client needed to connect to the cluster.
We start on Getting Access on how to request an account.
Once the access is granted, we proceed to Connect to the cluster (SSH) to install and use a SSH client that allows you to get a terminal on the HPC cluster.
A Terminal and its command line interface can be the first barrier for a user only familiar with Graphical User Interfaces (GUI), on UNIX/Linux Command Line Interface we present the most basic commands that will help you to create and manipulate folders and files.
The next step is to learn how to edit files, on Terminal-based Text Editors (nano) we show how to use nano a very simple but effective text editor.
We will use it to create the first program in FORTRAN that we will use later to explain execution on a batch system.
The next section, Data Storage, explains the several options to store data on the cluster, this is very important for first-time users as they have the tendency to rely only on the $HOME folder a storage space that is very limited in most scientific purposes.
At this point, we are ready to submit our first job. The section Workload Manager (SLURM) shows how to write a submission script and submit the job to the queue system.
Finally, section File Transfer (Globus) explains how to move files in and out between the HPC cluster and your own computer.
At the end of this tutorial, you should have your account activated, connected to spruce, able to create files and folders, submit simple jobs to the queue system and take the data back to your own machine.