Struct InfraPolicy 

pub struct InfraPolicy {
    pub max_signal_age_ms: u64,
    pub min_waiting_requests: u64,
    pub max_capable_inflight: u64,
}

Queue and capacity limits for moving requests to the capable model.

Fields

max_signal_age_ms: u64

Maximum age of the pool observation.

min_waiting_requests: u64

Every efficient worker must have at least this many waiting requests.

max_capable_inflight: u64

Stop moving requests when observed running or locally tracked capable work reaches this limit.

Implementations

impl InfraPolicy

pub fn validate(&self) -> Result<()>

Validate the policy once at configuration load.

pub fn apply( &self, outcome: &mut RoutingOutcome, models: &RuntimeModels, signals: &InfraSignals, ) -> Result<bool>

Adjust a decision-only task result. Returns true only when the model changes.

Hosts must serialize this check with their local in-flight increment, then decrement when the response ends or is cancelled. Worker selection remains in the serving layer.

Trait Implementations

impl Clone for InfraPolicy

fn clone(&self) -> InfraPolicy

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for InfraPolicy

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for InfraPolicy

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.