2 SFF machines, 1 Lenovo with Win10, 1 HP with Debian, both kick users off the local switch when asleep/frozen.

I have an endpoint with a handful of users connected to an unmanaged Netgear-ish switch, of the GS108 variety. These 2 machines I mentioned are business units with i5 CPUs and about 8GB of RAM, for whatever reason, will kick everyone else off the network ONLY when (A) The Windows machine goes into a deep sleep. Generally reproducible. (B) The debian machine freezes/becomes completely unresponsive (Used for signage, was using incognito firefox with Grafana, happened once so far.)

1. Has anyone had this happen? I cannot find similar examples online, numerous issues with Windows Sleep itself clogs the results.
2. Is this more likely a BIOS issue or a Power Supply Unit issue or a unmanaged switch issue?

My instinct feels like the managed switch up the chain is just ignoring all requests made when the offending machine sleeps and, if I may, shits out a bunch of requests or broadcasts or multicasts. I don't have time to trouble-shoot when it has occurred, and it is rare to happen, because getting those users back up is more valuable.

I do not think spending a lot of time on this is valuable but the obscure nature of the situation strikes my curiosity. If it can be solved, great.