From blocks to files
A storage device, whether a spinning disk, a solid-state drive or a memory card, does not know anything about documents or photos. At the lowest level it is a long run of numbered storage units. A file system is the scheme the operating system uses to turn that raw space into something people can use: named files, grouped in folders, with information about each one.
When you save a document, the file system finds free space, writes the content, and records where that content lives. When you open the file later, it uses that record to find the data again. Without such a scheme, the computer would have to remember the position of every piece of data itself.
Files and folders
A file is a named collection of data. The contents might be text, a spreadsheet, a program or a video. The file system does not usually care what the contents mean; that is up to the applications. Many systems use a file name extension, the letters after the final dot, as a hint about the type.
Folders (also called directories) hold files and other folders. Because folders can contain folders, the result is a tree. The very top is the root, and every file has a path: the route from the top, or from your current position, down to the file. The command line uses these paths directly, which is why they matter even for people who mostly click.
Folders are mainly an organizing aid for humans. A sensible structure, with clear names and consistent grouping, makes data easier to find, back up and clean up later.
Metadata: information about the data
Besides its contents, each file has metadata kept by the file system. This typically includes:
- the file name and its size
- dates, such as when it was created or last changed
- the owner
- the permissions that control who may do what
- where on the device the contents are stored
Metadata is why a folder listing can show sizes and dates without opening every file. It is also why moving a file within the same file system can be nearly instant: often only the record changes, not the data itself.
Permissions
On a shared computer or a server, not everyone should see or change everything. File systems that support permissions attach rules to files and folders. A common model distinguishes three basic actions:
- Read: view the contents.
- Write: change or delete the contents.
- Execute: run the file as a program, or, for a folder, pass through it to reach what is inside.
Rules are then given to the owner, to groups of users, and to everyone else. Some systems use more detailed lists of who is allowed which actions. The principle is the same in each case: a person gets the access that their work needs, and nothing more. This follows from least privilege, and in larger organizations it is usually managed through groups or roles, a topic related to role-based access, rather than file by file.
Permissions are not a substitute for data encryption. They are enforced by the operating system, so someone who can read the raw device by other means may bypass them. Encryption protects the content itself.
Formatting and file system types
Before a drive can store files, it is formatted with a file system. Formatting prepares the structures the system needs. Quick formatting often only sets up these structures and does not securely erase old data, which is worth knowing before disposing of a drive.
There are several file system types. Each operating system family has a favorite, and others are designed for particular uses, such as removable drives or very large volumes. They differ in:
- Compatibility. A drive formatted for one system may not be readable by another without extra software.
- Limits. Maximum file size and volume size vary.
- Features. Some support permissions, compression, encryption or snapshots, and some do not.
- Resilience. Many keep a journal, a log of intended changes, so that a sudden power loss is less likely to corrupt data.
The major desktop operating systems each have their own preferred format. When choosing a format for a shared external drive, compatibility is often the deciding factor.
Fragmentation, free space and performance
Over time, files are created, grown, deleted and rewritten. On a traditional spinning disk, a file may end up in scattered pieces, which can slow reading because the drive head has to move around. Solid-state drives have no moving parts, so scattering matters much less, though their own behavior and wear are managed differently.
Regardless of drive type, a nearly full volume tends to cause problems. The system needs working room, and applications may fail when they cannot write. Many administrators set up system monitoring to warn them well before storage runs out.
Mounting, shares and remote storage
Storage does not have to be a drive inside the machine. A file system can be attached, or mounted, from a USB stick, from a network share on a server, or from remote cloud storage presented as a folder. To the user it looks like another location in the tree, but speed, availability and permission handling can differ, since the data may be traveling over a network.
When things go wrong
Common problems include:
- Permission denied. Usually the account lacks the needed access, and the fix is a change of permission or group, not a workaround.
- Disk full. Free space by removing or archiving data, rather than deleting system files.
- Corruption after a crash or sudden removal. Eject drives properly. Most systems offer a repair tool, but it cannot always recover everything.
- Accidentally deleted files. Recovery may be possible but is never guaranteed, which is why a sound backup strategy matters.
Reading system logs often reveals the underlying reason for storage errors, such as a failing device or a full volume.
The short version
A file system gives raw storage structure: names, folders, ownership, access rules and records of where everything is. Understanding that structure helps with everyday tasks like finding data and sharing it safely, and with deeper tasks such as planning storage, restoring files and judging which drives will work between systems.