[Misc]: The meaning and usage of the parameter max_cache_size

[pyxyzc]

Your current environment

The output of above commands

I found the param max_cache_size in quickstart doc. My understanding is that users can use this parameter to set the total size of the underlying storage. However, after reading the code of nfsstore, I didn’t find any logic where it reads this parameter passed down from the inference engine, so I believe it is currently meaningless. Therefore, there are three usage-related questions:

1. Do other storage backend implementations accept this parameter? Furthermore, can the definitions of such parameter handling be organized and documented?
2. How does the underlying storage (taking NFS as an example) currently control or determine its initialized storage size?
3. I still believe exposing this parameter to the inference engine layer is meaningful — will it be re-enabled in the nfsstore implementation?

How would you like to use ucm.

I want to run inference of a [specific model](put link here). I don't know how to integrate it with unified.

[harrisonyhq]

The param max_cache_size currently only exits in dram connector

[pyxyzc]

The param max_cache_size currently only exits in dram connector

Thx!