Struct mz_dataflow_types::client::partitioned::PartitionedComputeState

pub struct PartitionedComputeState<T> {
    parts: usize,
    uppers: HashMap<GlobalId, MutableAntichain<T>>,
    peek_responses: HashMap<Uuid, HashMap<usize, PeekResponse>>,
    pending_tails: HashMap<GlobalId, Option<(MutableAntichain<T>, Vec<(T, Row, Diff)>)>>,
}

Maintained state for partitioned compute clients.

This helper type unifies the responses of multiple partitioned workers in order to present as a single worker.

Fields

parts: usize

Number of partitions the state machine represents.

uppers: HashMap<GlobalId, MutableAntichain<T>>

Upper frontiers for indexes and sinks.

peek_responses: HashMap<Uuid, HashMap<usize, PeekResponse>>

Pending responses for a peek; returnable once all are available.

pending_tails: HashMap<GlobalId, Option<(MutableAntichain<T>, Vec<(T, Row, Diff)>)>>

Tracks in-progress TAILs, and the stashed rows we are holding back until their timestamps are complete.

Implementations

impl<T> PartitionedComputeState<T> where
    T: Timestamp + Copy, 

fn reset(&mut self)

pub fn observe_command(&mut self, command: &ComputeCommand<T>)

Observes commands that move past, and prepares state for responses.

In particular, this method installs and removes upper frontier maintenance.

Trait Implementations

impl<T: Debug> Debug for PartitionedComputeState<T>

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

Formats the value using the given formatter.

impl<T> PartitionedState<ComputeCommand<T>, ComputeResponse<T>> for PartitionedComputeState<T> where
    T: Timestamp + Copy, 

fn split_command(
    &mut self,
    command: ComputeCommand<T>
) -> Vec<ComputeCommand<T>>

Splits a command into multiple partitions.

fn absorb_response(
    &mut self,
    shard_id: usize,
    message: ComputeResponse<T>
) -> Option<Result<ComputeResponse<T>, Error>>

Absorbs a response from a single partition.