Snapshot

A snapshot is a point-in-time record of an operating system, disk or virtual machine that lets you instantly restore exactly that state if something goes wrong. A snapshot is not a full copy of your data — it is more of a checkpoint you can return to when an update, deployment or configuration change misbehaves.

How a snapshot works

Most file systems and virtualization platforms create snapshots using copy-on-write. At the moment of creation the system freezes the current view of the data, and from then on every new write is stored in a separate area while the original blocks stay untouched. As a result, taking a snapshot completes in a fraction of a second and initially consumes almost no space — it only grows as more blocks are modified.

Rolling back to a snapshot discards every change made after it was taken and restores the frozen state. Keep in mind that a snapshot shares the storage medium with your production data, so a disk failure destroys both — which is why a snapshot complements but never replaces a full backup.

Snapshots in practice

Snapshots are invaluable before risky operations: ahead of a system update, a database migration or a new application release, an administrator takes a snapshot and, if trouble appears, reverts to it in seconds instead of rebuilding the whole environment from a backup. On VPS servers the control panel typically lets you snapshot a machine with a single click — a convenient safety net for testing and configuration. Snapshots are also used as a consistent starting point for backups: the system takes a snapshot and then calmly copies data from it to external storage without blocking the running application.