Block Storage vs Object Storage
People think object storage is just a disk in the cloud, and it is not. Block storage is a raw disk that one server uses as its own drive, which suits databases. Object storage is a huge bucket of files reached over the web, each with a name. You cannot run a database on it. But it is cheap and holds as many files as you like, which suits backups.
Block Storage
Raw disk space a server uses as its own drive. Fast, and good for databases.
Block storage is raw disk space that a machine sees as its own drive. The space is split into fixed blocks, and the machine's own system decides what goes in each one. It is fast, so it suits operating systems and databases. In the cloud, a block volume is most often attached to one server at a time, like plugging in a disk.
Object Storage
A cheap store where each file is saved whole and found by its key.
Object storage keeps each file as an object with its own key, inside a container called a bucket. There are no real folders, so a slash in the key is just part of its name. You save, read or replace a whole object over the web, but you cannot edit part of one. It is cheap and grows with no limit to plan for. That is why data lakes are built on it.
In one line
- Block Storage
- Raw disk space a server uses as its own drive. Fast, and good for databases.
- Object Storage
- A cheap store where each file is saved whole and found by its key.
These live in different corners of the glossary (Infrastructure and Data), which is exactly why they get confused.