Unlocking Linux Essentials: From GNU/GPL to RAID and Filesystem Basics

1. GNU and GPL

GNU is the GNU Project (also known as the “GNU Project”), launched publicly by Richard Stallman on September 27, 1983. Its goal is to create a completely free operating system; GNU is also called the free software engineering project.

GPL is the GNU General Public License, a “copyleft” license that protects GNU software so it can be freely used, copied, studied, modified, and distributed, while requiring that the software be released in source‑code form.

When combined with the Linux kernel, GNU forms a complete operating system often referred to as “GNU/Linux” or simply Linux.

2. Linux Distributions

A typical Linux distribution includes the Linux kernel, GNU libraries and tools, a command‑line shell, the X Window System with a desktop environment such as KDE or GNOME, and thousands of applications ranging from office suites to compilers and scientific tools.

Mainstream distributions: Red Hat Enterprise Linux, CentOS, SUSE, Ubuntu, Debian, Fedora, Gentoo.

3. Unix and Linux

Linux is based on Unix, belonging to the Unix family. Unix systems support multiple users, multitasking, multithreading, and various CPU architectures. Linux inherits Unix’s network‑centric design and provides a stable, multi‑user network operating system.

4. Swap Partition

Swap (the swap partition) is used when physical memory is insufficient: the system moves inactive pages to disk space, freeing RAM for active processes. When needed, data is swapped back into memory.

Swap space should be at least equal to the size of physical memory, not less than 64 MiB, and at most twice the RAM size.

5. GRUB Concept

GNU GRUB (Grand Unified Bootloader) is a multi‑OS boot manager from the GNU project. It allows a computer with several operating systems to select which one to boot, can load different Linux kernels, and can pass boot parameters such as entering single‑user mode.

6. Buffer and Cache

Cache (缓存) is a small, fast storage area between the CPU and main memory. By caching file data blocks, it mitigates the speed mismatch between CPU processing and memory I/O, accelerating data exchange; larger caches yield faster CPU performance.

Buffer (缓冲) is a high‑speed buffer that caches disk (I/O device) data blocks, speeding up disk access and reducing I/O. Buffers hold data about to be written to disk, whereas caches hold data read from disk.

7. TCP Three‑Way Handshake

Client sends a SYN packet (SYN=A) and waits for a response.

Server receives SYN and replies with SYN (A+1) plus its ACK (K).

Client receives the SYN+ACK and sends an ACK (K+1) back.

After this exchange the TCP connection is established and data transmission can begin.

8. Linux Filesystem Hierarchy

Linux uses a single‑root tree structure. Key directories include:

/ – root of the entire filesystem.

/boot – kernel and boot loader files (e.g., kernel, initrd, GRUB).

/bin – essential user commands (similar to /usr/bin).

/sbin – system administration commands (root only).

/etc – configuration files.

/dev – device files (terminals, disks, optical drives, etc.).

/var – variable data such as logs and mail.

/home – default home directories for regular users.

/opt – third‑party software packages.

/lib – shared libraries and kernel modules.

9. Hard and Soft Links

Hard link (Hard Link) uses the same inode number, allowing multiple filenames to point to the same file. Deleting one hard link does not affect the underlying file or its other links. ln source new-link Soft link (Symbolic Link) creates a path‑based shortcut, similar to a Windows shortcut. Multiple filenames can point to the same source file; if the source is removed, all soft links become invalid. Soft links support directories and can cross partitions or filesystems.

ln -s source new-link

10. RAID Technology

RAID (Redundant Arrays of Independent Disks) combines multiple physical disks into a logical volume, offering higher performance and data redundancy.

RAID groups disks to provide faster access, fault tolerance, and various trade‑offs between capacity, speed, and reliability.

RAID levels:

RAID 0 – striping; parallel read/write across disks, no redundancy, high throughput.

RAID 1 – mirroring; duplicate data on two disks for redundancy, improves read speed, slower writes, supports hot‑swap.

RAID 0+1 (RAID 10) – combines striping and mirroring; requires at least four disks, provides both performance and redundancy.

RAID 5 – striping with parity; balances performance, data safety, and storage efficiency, needs at least three disks.