User guide: Machine placement
How spec.placement steers which node a Machine lands on, and what today's
User guide: topology-aware placement (spec.topology, spec.devices)
Places a Machine's vCPUs, memory and PCI devices by host locality: NUMA
User guide: accelerator profiles (MachineAcceleratorProfile, experimental)
A MachineAcceleratorProfile is a named accelerator shape that a cluster
User guide: vGPU profiles, MIG, GPU memory and GPU health (experimental)
A Machine can ask for a slice of a GPU instead of a whole one. The slice
User guide: SR-IOV NIC passthrough via DRA/VFIO reuse
A dedicated, high-performance network interface passed straight through
User guide: node maintenance mode and DRS rebalancing
Two kairon-controller features keep Machines where they should be:
Preemption
spec.priority still orders pending Machines. It halts a running Machine only when both sides opt in.
User guide: Mac Fabric (experimental)
Mac Fabric turns several Apple silicon Macs into one pool of vz Machines.