Enum MultiRunnerClient 

pub enum MultiRunnerClient<B1: BackendIr, B2: BackendIr, B3: BackendIr, B4: BackendIr> {
    B1(Runner<B1>),
    B2(Runner<B2>),
    B3(Runner<B3>),
    B4(Runner<B4>),
}

A local client with multiple runners (each responsible to execute tensor operations on a given backend).

Variants

B1(Runner<B1>)

B2(Runner<B2>)

B3(Runner<B3>)

B4(Runner<B4>)

Trait Implementations

impl<B1: Clone + BackendIr, B2: Clone + BackendIr, B3: Clone + BackendIr, B4: Clone + BackendIr> Clone for MultiRunnerClient<B1, B2, B3, B4>

fn clone(&self) -> MultiRunnerClient<B1, B2, B3, B4>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<B1: BackendIr, B2: BackendIr, B3: BackendIr, B4: BackendIr> RunnerClient for MultiRunnerClient<B1, B2, B3, B4>

type Device = MultiDevice<B1, B2, B3, B4>

Device type.

fn register_op(&self, op: OperationIr)

Register a new tensor operation to be executed by the (runner) server.

fn read_tensor_async( &self, tensor: TensorIr, ) -> DynFut<Result<TensorData, ExecutionError>>

Read the values contained by a tensor.

fn register_tensor_data(&self, data: TensorData) -> RouterTensor<Self>

Create a new RouterTensor from the tensor data.

fn device(&self) -> Self::Device

Get the current device used by all operations handled by this client.

fn sync(&self) -> Result<(), ExecutionError>

Sync the runner, ensure that all computations are finished.

fn seed(&self, seed: u64)

Seed the runner.

fn create_empty_handle(&self) -> TensorId

Create a new (uninitialized) empty tensor and returns its corresponding tensor id.

fn supports_dtype(&self, dtype: DType) -> bool

Whether the type is supported.

fn register(&self, op: OperationIr) -> Vec<RouterTensor<Self>>

Register a new tensor operation to be executed by the (runner) server.