dynamo.nixl_connect.WriteOperation#
An operation which transfers data from the local worker to a remote worker.
To create the operation, RDMA metadata (RdmaMetadata) from a remote worker’s WritableOperation
along with a matching set of local Descriptor objects which reference memory to be transferred to the remote worker must be provided.
The RDMA metadata must be transferred from the remote to the local worker via a secondary channel, most likely HTTP or TCP+NATS.
Once created, data transfer will begin immediately. Disposal of the object will instruct the RDMA subsystem to cancel the operation, therefore the operation should be awaited until completed unless cancellation is intended. Cancellation is handled asynchronously.
Example Usage#
async def write_to_remote(
self,
remote_metadata: dynamo.nixl_connect.RdmaMetadata,
local_tensor: torch.Tensor
) -> None:
descriptor = dynamo.nixl_connect.Descriptor(local_tensor)
with self.connector.begin_write(descriptor, remote_metadata) as write_op:
# Wait for the operation to complete writing local_tensor to the remote worker.
await write_op.wait_for_completion()
Methods#
cancel#
def cancel(self) -> None:
Instructs the RDMA subsystem to cancel the operation. Completed operations cannot be cancelled.
wait_for_completion#
async def wait_for_completion(self) -> None:
Blocks the caller until all provided buffers have been transferred to the remote worker.
Properties#
status#
@property
def status(self) -> OperationStatus:
Returns OperationStatus which provides the current state (aka. status) of the operation.