> For the complete documentation index, see [llms.txt](https://isubasinghe.gitbook.io/isithas-wiki/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://isubasinghe.gitbook.io/isithas-wiki/computer_science/distributed_systems/distributed_file_systems.md).

# distributed\_file\_systems

## Basic idea

File systems whose data and metadata are spread across many networked machines while still presenting a unified namespace and POSIX-ish semantics (or a deliberately weaker model). Hard because performance, consistency, and fault-tolerance pull against each other.

## Key facts

* Examples: GFS, HDFS, Ceph, Lustre, AFS, NFS.
* Replication factor $r$ tolerates $r-1$ node failures.
* Consistency models range from strict (linearisable) to close-to-open (NFS) to eventually consistent.