How and when to use /dev/shm for efficiency?
filesystems, io, linux, performance, shared-memory
Solution
You don't use `/dev/shm`. It exists so that the POSIX C library can provide shared memory support via the POSIX API. Not so you can poke at stuff in there.
If you want an in-memory filesystem of your very own, you can mount one wherever you want it.
`mount -t tmpfs tmpfs /mnt/tmp`, for example.
A Linux `tmpfs` is a temporary filesystem that only exists in RAM. It is implemented by having a file cache without any disk storage behind it. It will write its contents into the swap file under memory pressure. If you didn't want the swapfile you can use a `ramfs`.
I don't know where you got the idea of using `/dev/shm` for efficiency in reading files, because that isn't what it does at all.
Maybe you were thinking of using memory mapping, via the `mmap` system call?
Read this answer here: https://superuser.com/a/1030777/4642 it covers a lot of tmpfs information.
Problem
How is `/dev/shm` more efficient than writing the file on the regular file system? As far as I know, `/dev/shm` is also a space on the HDD so the read/write speeds are the same. My problem is, I have a 96GB file and only 64GB RAM (+ 64GB swap). Then, multiple threads from the same process need to read small random chunks of the file (about 1.5MB). Is `/dev/shm` a good use case for this? Will it be faster than opening the file in read-only mode from `/home` and then passing over to the threads to do the reading the required random chunks?