> The primary use case in mind for parker is on the machines with high core counts, where scalability concerns may arise. Once started, there is no communication between kernel instances. In other words, they share nothing thus improve scalability.
Curious what benefits this would have vs the usual virtualization? VM’s are the typical strategy for subdividing machines with high core counts into multiple kernels, and seems to work well enough?
This reminds me of https://barrelfish.org/
Ok, but why?
From TFA:
> The primary use case in mind for parker is on the machines with high core counts, where scalability concerns may arise. Once started, there is no communication between kernel instances. In other words, they share nothing thus improve scalability.
Curious what benefits this would have vs the usual virtualization? VM’s are the typical strategy for subdividing machines with high core counts into multiple kernels, and seems to work well enough?
Think it can be used for lockstep computing.
Also seems to be a way to bring microkernel benefit to Linux. Isolate system services.