pub struct ScheduledRequestMetrics {
pub num_prefill_requests: u32,
pub sum_prefill_tokens: u32,
pub var_prefill_length: f64,
pub sum_prefill_kv_tokens: u32,
pub num_decode_requests: u32,
pub sum_decode_kv_tokens: u32,
pub var_decode_kv_tokens: f64,
}Expand description
Metrics for requests scheduled in one forward-pass iteration.
This mirrors Dynamo ForwardPassMetrics v1 scheduled request telemetry. AIC owns this Rust copy so AIC does not depend on Dynamo crates.
Fields§
§num_prefill_requests: u32Number of prefill requests, including new requests and chunked-prefill continuations.
sum_prefill_tokens: u32Total tokens freshly computed for prefill requests in this iteration.
var_prefill_length: f64Population variance of total prompt lengths across prefill requests.
sum_prefill_kv_tokens: u32Total KV tokens read for prefill requests, including prefix cache hits and previously computed chunks.
num_decode_requests: u32Number of decode requests.
sum_decode_kv_tokens: u32Total KV context length across decode requests.
var_decode_kv_tokens: f64Population variance of KV context lengths across decode requests.
Trait Implementations§
Source§impl Clone for ScheduledRequestMetrics
impl Clone for ScheduledRequestMetrics
Source§fn clone(&self) -> ScheduledRequestMetrics
fn clone(&self) -> ScheduledRequestMetrics
Returns a duplicate of the value. Read more
1.0.0 (const: unstable) · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
Performs copy-assignment from
source. Read moreSource§impl Debug for ScheduledRequestMetrics
impl Debug for ScheduledRequestMetrics
Source§impl Default for ScheduledRequestMetrics
impl Default for ScheduledRequestMetrics
Source§fn default() -> ScheduledRequestMetrics
fn default() -> ScheduledRequestMetrics
Returns the “default value” for a type. Read more
Source§impl<'de> Deserialize<'de> for ScheduledRequestMetrics
impl<'de> Deserialize<'de> for ScheduledRequestMetrics
Source§fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
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 PartialEq for ScheduledRequestMetrics
impl PartialEq for ScheduledRequestMetrics
Source§impl Serialize for ScheduledRequestMetrics
impl Serialize for ScheduledRequestMetrics
impl StructuralPartialEq for ScheduledRequestMetrics
Auto Trait Implementations§
impl Freeze for ScheduledRequestMetrics
impl RefUnwindSafe for ScheduledRequestMetrics
impl Send for ScheduledRequestMetrics
impl Sync for ScheduledRequestMetrics
impl Unpin for ScheduledRequestMetrics
impl UnsafeUnpin for ScheduledRequestMetrics
impl UnwindSafe for ScheduledRequestMetrics
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more