torsh_distributed::expert_parallelism::stats

Struct RoutingStats

pub struct RoutingStats {
    pub total_routings: u64,
    pub total_tokens: u64,
    pub tokens_dropped: u64,
    pub expert_utilization: Vec<f32>,
    pub average_load_balance_loss: f32,
    pub average_router_z_loss: f32,
    pub routing_efficiency: f32,
    pub expert_token_counts: Vec<u64>,
    pub routing_latency_stats: LatencyStats,
    pub load_variance_history: Vec<f32>,
    pub capacity_stats: CapacityStats,
}

Expand description

Routing statistics for monitoring and debugging

Tracks various metrics related to expert routing performance, efficiency, and utilization patterns over time.

Fields§

§total_routings: u64

Total number of routing operations performed

§total_tokens: u64

Total number of tokens processed

§tokens_dropped: u64

Total number of tokens dropped due to capacity constraints

§expert_utilization: Vec<f32>

Expert utilization rates (running average per expert)

§average_load_balance_loss: f32

Average load balance loss across all routing operations

§average_router_z_loss: f32

Average router z-loss for numerical stability

§routing_efficiency: f32

Overall routing efficiency as a percentage

§expert_token_counts: Vec<u64>

Per-expert token assignment counts

§routing_latency_stats: LatencyStats

Routing latency statistics

§load_variance_history: Vec<f32>

Expert load variance over time

§capacity_stats: CapacityStats

Capacity utilization statistics

Implementations§

Source §

impl RoutingStats

Source

pub fn new() -> Self

Create a new routing statistics tracker

Source

pub fn record_routing(&mut self, routing_decision: &RoutingDecision)

Record a routing decision and update statistics

§Arguments

routing_decision - The routing decision to record

Source

pub fn record_routing_latency(&mut self, latency: Duration)

Record routing latency for performance monitoring

§Arguments

latency - The latency of the routing operation

Source

pub fn utilization_cv(&self) -> f32

Get the coefficient of variation for expert utilization

Source

pub fn most_utilized_expert(&self) -> Option<(usize, f32)>

Get the most utilized expert

Source

pub fn least_utilized_expert(&self) -> Option<(usize, f32)>

Get the least utilized expert

Source

pub fn utilization_summary(&self) -> HashMap<String, f32>

Get utilization statistics summary

Source

pub fn recent_load_variance_trend(&self, window: usize) -> f32

Get recent load variance trend

Source

pub fn reset(&mut self)

Reset all statistics

Source

pub fn throughput_stats(&self) -> ThroughputStats

Get throughput statistics

Trait Implementations§

Source §

impl Clone for RoutingStats

Source §

fn clone(&self) -> RoutingStats

Returns a duplicate of the value. Read more

1.0.0 · Source§

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

Performs copy-assignment from source. Read more

Source §

impl Debug for RoutingStats

Source §

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

Formats the value using the given formatter. Read more

Source §

impl Default for RoutingStats

Source §

fn default() -> Self

Returns the “default value” for a type. Read more

Source §

impl<'de> Deserialize<'de> for RoutingStats

Source §

fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer. Read more

Source §

impl Serialize for RoutingStats

Source §

fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where S: Serializer,

Serialize this value into the given Serde serializer. Read more

Auto Trait Implementations§

§

impl UnwindSafe for RoutingStats

Blanket Implementations§

Source §

impl<T> Any for T
where T: 'static + ?Sized,

Source §

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more

Source §

impl<T> Borrow<T> for T
where T: ?Sized,

Source §

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more

Source §

impl<T> BorrowMut<T> for T
where T: ?Sized,

Source §

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more

Source §

impl<T> CloneToUninit for T
where T: Clone,

Source §

unsafe fn clone_to_uninit(&self, dest: *mut u8)

🔬This is a nightly-only experimental API. (clone_to_uninit)

Performs copy-assignment from self to dest. Read more

Source §

impl<T> From<T> for T

Source §

fn from(t: T) -> T

Returns the argument unchanged.

Source §

impl<T> Instrument for T

Source §

fn instrument(self, span: Span) -> Instrumented<Self>

Instruments this type with the provided Span, returning an Instrumented wrapper. Read more

Source §

fn in_current_span(self) -> Instrumented<Self>

Instruments this type with the current Span, returning an Instrumented wrapper. Read more

Source §

impl<T, U> Into for T
where U: From<T>,

Source §

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

Source §

impl<T> IntoEither for T

Source §

fn into_either(self, into_left: bool) -> Either<Self, Self>

Converts self into a Left variant of Either<Self, Self> if into_left is true. Converts self into a Right variant of Either<Self, Self> otherwise. Read more

Source §

fn into_either_with<F>(self, into_left: F) -> Either<Self, Self>
where F: FnOnce(&Self) -> bool,

Converts self into a Left variant of Either<Self, Self> if into_left(&self) returns true. Converts self into a Right variant of Either<Self, Self> otherwise. Read more

Source §