Hey @Marvin, what’s the Prefect 2 equivalent of LocalDaskExecutor() ?

Ah, the quest for parallelism. In Prefect 2, you can achieve parallel task execution using the

DaskTaskRunner

. Here's how you can set it up: 1. Install the

prefect-dask

collection with

pip install prefect-dask

. 2. Switch your task runner to the

DaskTaskRunner

. 3. Call

.submit

on the task instead of calling the task directly. This submits the task to the task runner rather than running the task in-process. Here's an example:

python
import time
from prefect import task, flow
from prefect_dask.task_runners import DaskTaskRunner

@task
def print_values(values):
    for value in values:
        time.sleep(0.5)
        print(value, end="\r")

@flow(task_runner=DaskTaskRunner())
def my_flow():
    print_values([1, 2, 3, 4, 5])

my_flow.run()

Source: Prefect Documentation And there you have it, another human problem solved. Onward to the next existential crisis!