Principles of file indexing in Linux I have read a lot about indexing in Mysql and I know that Linux has a great utility `locate` to find files. Maybe Linux has the same principle to do indexing as Mysql have provided. But I think the nugget of indexing lies in the structure of `inodes` in Linux. My question is: What are the best ways of implementing indexing in Linux? (e.g. directory index, file index...) And how indexing works in most usual cases? I would be really appreciative for the full answer. **P.S** I have already cut through `stackoverflow`, `unix.stackexchange` and `askubuntu` trying to find answer: indexing, support indexing and so on.

There is no "mechanism of any file indexing" in Linux kernel.

There are some user-space tools such as `locate` (+updatedb), KDE Nepomuk and similar.

Now, filesystems may have use some clever techniques to store directory indexes, but it has nothing to do with locate & friends - it is merely a filesystem implementation detail, and it depends on what filesystem is being used.