Linux Basics¶

Every HPC cluster runs Linux. When you SSH into Discovery, you're dropped into a Linux shell: a text-based interface where you control the system by typing commands. There's no file browser, no right-click menu, and you can't use your mouse. Just a prompt, a blinking cursor, and a lot of power.

If you've never used the command line before, this can feel disorienting. This article will give you the vocabulary and hands-on practice you need to navigate confidently.

The Terminal Tour¶

You've just been given access to Discovery. The previous grad student left behind some data on a lab volume. Your mission: find it, understand it, and set up your own workspace.

Enter your NetID below to personalize the simulation, then click SSH into Discovery → to begin.

The Shell Prompt¶

When you SSH into Discovery, you see something like this:

[f00abc@discovery ~]$

This is your shell prompt. It tells you:

• f00abc — your username (your NetID)
• discovery — the name of the host you're logged into
• ~ — your current directory (~ is shorthand for your home directory)
• $ — you're logged in as a regular user (not root)

Everything you type after $ is a command. Press Enter to run it.

Navigation¶

Where am I? — pwd¶

pwd (print working directory) tells you your current location in the filesystem:

pwd

/dartfs/rc/home/c/f00abc

What's here? — ls¶

ls lists the contents of your current directory:

ls

Add flags to get more detail:

Move around — cd¶

cd (change directory) moves you to a new location:

cd /dartfs/rc/lab/C/ChenLab   # absolute path
cd my_project                  # relative path (inside current dir)
cd ..                          # go up one level
cd ~                           # go home
cd                             # also goes home

Reading Files¶

Print a file — cat¶

cat README.txt

cat prints the entire file to the terminal. Fine for small files; overwhelming for large ones.

Peek at the top — head¶

head results.csv        # first 10 lines (default)
head -n 50 results.csv  # first 50 lines

Peek at the bottom — tail¶

tail results.csv         # last 10 lines
tail -n 20 results.csv   # last 20 lines
tail -f slurm-12345.out  # follow a file as it grows (useful for job logs)

Page through a file — less¶

less results.csv

less opens a scrollable viewer. Use Up / Down to scroll, Space to page down, q to quit, /pattern to search.

File Management¶

Create a directory — mkdir¶

mkdir my_project
mkdir -p my_project/data/raw   # create nested directories at once

Copy a file — cp¶

cp source.txt destination.txt          # copy file
cp source.txt /some/other/directory/   # copy into a directory
cp -r my_dir/ backup_dir/              # copy a whole directory (-r = recursive)

Move or rename — mv¶

mv old_name.txt new_name.txt           # rename
mv file.txt /dartfs/rc/lab/C/ChenLab/ # move to another directory

Delete — rm¶

rm old_file.txt
rm -r old_directory/   # remove a directory and everything inside it

Searching¶

Search file contents — grep¶

grep searches for a pattern inside files:

grep "PASS" results.csv           # lines containing "PASS"
grep -i "pass" results.csv        # case-insensitive
grep -n "error" slurm-12345.out   # show line numbers
grep -r "TODO" my_project/        # search recursively through a directory

Find files — find¶

find . -name "*.csv"              # find all CSV files in current directory
find /dartfs/rc/lab -name "*.bam" # find BAM files in the lab volume
find . -newer reference.txt       # files newer than reference.txt

Getting Help¶

Every command has built-in documentation:

man ls        # full manual page — press q to quit
ls --help     # shorter usage summary

When in doubt, man is your friend.

Paths and the Filesystem¶

Linux uses a single unified filesystem tree rooted at /. There are no drive letters — everything hangs off /.

/
├── dartfs/
│   └── rc/
│       ├── home/
│       │   └── c/
│       │       └── f00abc/      ← your home directory (~)
│       └── lab/
│           └── C/
│               └── ChenLab/     ← a lab storage volume
└── scratch/                     ← fast scratch storage

Absolute vs. relative paths¶

An absolute path starts from / and unambiguously identifies a location regardless of where you currently are:

cd /dartfs/rc/lab/C/ChenLab
cat /dartfs/rc/home/c/f00abc/README.txt

A relative path is relative to your current directory. If you're already in your home directory:

cat README.txt          # same file, shorter to type
cd ../..                # go up two levels

Special path shortcuts¶

Quick Reference¶

What's Next¶

With these fundamentals in hand, you're ready to go deeper:

• Linux — Permissions, Pipes & the Environment — file permissions, pipes, redirection, and shell configuration
• Storage on Discovery — understand your home directory, lab volumes, and scratch space
• Submitting your first job — put the scheduler to work with sbatch