Skip to content

Linux Basics

Every HPC cluster runs Linux. When you SSH into Unity, 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.

Interactive first

Before reading the reference sections below, try the hands-on terminal tour. You'll practice the most important commands in a guided, story-driven simulation without needing real cluster access.

The Terminal Tour

You've just been given access to Unity. 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 Unity → to begin.

The Shell Prompt

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

[username@login1 ~]$

This is your shell prompt. It tells you:

  • username: your username
  • login1: the name of the login node you're connected to
  • ~: 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.

Where am I? — pwd

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

pwd

/home/username

What's here? — ls

ls lists the contents of your current directory:

ls

Add flags to get more detail:

Command What it does
ls List files and directories
ls -l Long format; shows permissions, size, date
ls -a Show hidden files (those starting with .)
ls -lh Long format with human-readable file sizes
ls /some/path List a different directory without navigating to it

Move around — cd

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

cd /work/pi_chen       # absolute path
cd my_project                  # relative path (inside current dir)
cd ..                          # go up one level
cd ~                           # go home
cd                             # also goes home

Reading Files

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 /work/pi_chen/  # move to another directory

There's no Recycle Bin

On Linux, rm deletes immediately and permanently. There is no undo. Double-check your paths before running rm on a shared filesystem.

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 /work -name "*.bam"  # find BAM files in work storage
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 /.

/
├── home/
│   └── username/                ← your home directory (~)
├── work/
│   └── pi_chen/                 ← your PI group's work directory
├── scratch/                     ← fast temporary storage
├── project/                     ← allocated shared storage
└── datasets/                    ← curated read-only datasets

Absolute vs. relative paths

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

cd /work/pi_chen
cat /home/username/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

Shortcut Means
~ Your home directory
. Current directory
.. Parent directory
- Previous directory (cd - goes back)

Quick Reference

Category Command What it does
Navigation pwd Print current directory
ls List files
ls -la List all files, long format
cd /path Change directory
cd ~ Go home
cd .. Go up one level
Reading cat file Print whole file
head file First 10 lines
tail file Last 10 lines
less file Scrollable viewer
Files mkdir dir Create directory
cp src dst Copy file
mv src dst Move/rename
rm file Delete file (permanent!)
Search grep pattern file Search in file
find . -name "*.csv" Find files by name
Help man cmd Full manual
cmd --help Short help

What's Next

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