Introduction
|
Learn about CPUs, cores, and cache, and compare your machine with an HPC cluster
Identify how an HPC cluster could benefit your research.
|
Open On-Demand
|
Open On-demand allow you to create interactive sessions using Jupyter and RStudio
Open On-demand includes a basic text editor, file manager and the ability to visualize plots in JPEG, PDF and PNG
Open On-demand interacts with the queue system to monitor and control jobs running on the cluster.
|
Command Line Interface
|
The basic commands you must know are echo , cat , date , pwd , cd , mkdir , touch , cp , mv , rm . You will use these commands very often.
|
Adjurn
|
|
Terminal-based Text Editors
|
For editing files from the terminal there are several editor available
nano is an easy to use editor with commands that are shown on the bottom, good for beginners.
emacs is a full featured editor that relies on multiple key combinations to control its operation.
vi/vim is installed by default on every Linux system, it works by changing between ‘modes’
|
Environment Modules
|
Use module avail to know all the modules on the cluster.
Use module load <module_name> to load the module that you need.
You can preload modules for each login by adding the load line on your $HOME/.bashrc
|
Scheduling jobs
|
The scheduler handles how compute resources are shared between users.
Everything you do should be run through the scheduler.
A job is just a shell script.
If in doubt, request more resources than you will need.
It is a good idea to keep aliases to common torque commands for easy execution.
|
Adjurn
|
|
Terminal Multiplexing: tmux
|
|
Environment management: Conda
|
|
Software Containers: Singularity
|
|
Adjurn
|
|
Version Control: git
|
|
Compiling Codes
|
|
Shell Scripting
|
|
Final remarks
|
|