Unlocking Linux Storage: How Ext Filesystems, RAID, and LVM Work Together

In modern Linux systems the dominant file systems are Ext (2/3/4), which consist of a Super Block, an Inode table, and Data blocks.

Because storing all Inodes and Blocks together would be unwieldy, Ext formats the disk into multiple Block Groups, each containing its own independent Inode table, Data blocks, and a copy of the Super Block, similar to how a military division is composed of many brigades.

The Super Block records essential information about the entire file system; without it the file system cannot function. It stores:

the total number of Blocks and Inodes;

the counts of used and free Blocks/Inodes;

the size of Blocks (1 KB, 2 KB, 4 KB) and Inodes (128 bytes);

timestamps such as mount time, last write time, and last fsck time;

a valid‑bit indicating whether the file system is currently mounted (0) or not (1).

RAID devices are divided into hardware RAID and software RAID. Hardware RAID relies on a dedicated RAID controller card, offering high performance without consuming CPU or memory, but at a higher cost and requiring driver installation. Software RAID is implemented by the operating system, consuming CPU and I/O resources, but is inexpensive and easy to set up, making it suitable for budget‑constrained environments that still need data safety.

Linux provides the mdadm utility for software RAID, supporting common RAID levels and identifying hardware RAID as /dev/sd{a,b,c} and software RAID as /dev/md{0,1,2}.

Traditional partitioning fixes partition sizes, making space adjustments difficult. Logical Volume Manager (LVM) solves this by allowing dynamic resizing of file system capacity. LVM aggregates several physical partitions or disks into a Volume Group (VG), which can then be subdivided into Logical Volumes (LV) that are mounted like regular partitions. While LVM offers flexible capacity management, it does not improve performance or data redundancy; a failure can cause loss of all data stored on the LVM.