High-Throughput Extended File System

Why?

Picture this:

You're in a HFT firm, with your data center next to the NYSE or whatever location you're at, but you really want an edge, not do micro-optimizations to save some micro-seconds.

So what would you need to do?

Maybe you can pull off a file-system in the storage layer in order to speed-up virtual memory lookup or handle big inodes effectively?

How?

Well...

That's why I've started working on HeFS, for that kind of edge cases where high-throughput and correctness is needed.

Where is it?

Check the GitHub Project, where I'm currently implementing it!

And what language do you use for the implementation?

C++ (-Werror, -Wpedantic, -Wall) (and C++ 20)