[LSF/MM/BPF TOPIC] nvme-fabrics at scale

[Daniel Wagner]

We are working on the next major version of nvme-cli and libnvme. The 
goal is simple: make the library API more intuitive and fix the "growing 
pains" of our current design choices. This refactoring is essential for 
preparing the codebase for future features. So I'd like to give a quick 
update and gather feedback.

The current discovery/auto-connect process is a "black box" that is hard 
to debug and tends to struggle in large-scale environments. I like to go 
through these existing pain points and demonstrate a first PoC for an 
alternative approach that improves stability and observability.

Daniel