@JustinDerrick@infosec.exchange
Post #3291519
2026-06-02 14:08 UTC
@valkyrie@wandering.shop I don’t know enough about high performance computing / clustering to know if this type of network-based swap (or caching) is actually a thing, but it’s similar to Dynamic LPARs on IBM hardware/hypervisors, with the restriction of the host machine’s RAM.
It would be neat to have a cluster of SBC’s running on a gigabit switch, then a desktop making all of its memory available to the cluster. But is this re-inventing the wheel?
Replies (1)
-
@valkyrie@wandering.shop 2026-06-02 14:11
@JustinDerrick@infosec.exchange it might be. i know for me, when i think about this kind of solution and look around for one that isn't made for corporate and highly resourceful/advanced hardware that anyone can easily DIY, i don't see very much. that's what motivated me to make this proof of concept as a thought experiment on gift and library economies in federated computing. maybe it's an old tired idea, but it certainly isn't at the forefront of most design thought nowadays. source: https://github.com/solarpunkgirl/vesta