Basically, in all file systems, the first block is used for the Boot Sector and the second block is used for the Inodes (the table of pointers that localize the data in the disk). The rest of the dist is used for real data.
- EXT (4 is the latest version)
- The most popular and widely supported.
- Improves storage of large files because of the concept of “Extents”, which uses sequential blocks to prevent data fragmentation.
- Prevents data corruption and allows easier system recovery via journaling (32 bits).
- Uses buffering to prevent fragmentation and increase performance and also permits defragmentation while mounted.
- XFS (popular in enterprises)
- Journaling (64 bits) file system made for huge amounts of data and higher throughputs.
- The file system can be defragmented or resized online (without unmounting).
- ZFS (combined file system)
- Highest storage size that combines multiple physical storages in storage pools.
- Capable of journaling, checksum, snapshots, RAID-Z, and silent fix of corrupted data.
- Proprietary.
- Btrfs (high-performance copy on write)
- Not proprietary and probably the future next-generation industry standard due to its overall features.
- Similar to ZFS except for the max storage size but offers transparent compression and optimization for SSD.
- The copy on write does not overwrite data when modifying it. That way it prevents corruption in case of loss of power and other issues.
- Has a tool to convert EXT 2/3/4 to Brtfs.