Function current_numa_node 

pub fn current_numa_node() -> u32

Per-NUMA-node sidecar sharding scaffolding. The default global sidecar handles all instances; multi-sidecar deployment with per-NUMA pinning would spawn one Sidecar per node and route registrations by spawn-thread affinity. The detection function numa_node_count surfaces the topology; the routing layer plugs in here when load testing on a multi-socket host motivates it.

Uses GetCurrentProcessorNumberEx + GetNumaProcessorNodeEx on Windows: these two work across processor groups (Windows splits logical processors into groups of up to 64), so the >64-logical- processor case (dual-socket servers, large core-count workstations) is handled correctly. The legacy GetNumaProcessorNode (capped at processor 255) is no longer called.