Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

<@ULVA73B9P> How can I use the `.result()` function on tasks but run tasks concurrently

To run tasks concurrently and use the `.result()` function in Prefect, you can use task runners that facilitate parallel execution. Prefect provides several task runners like `ThreadPoolTaskRunner`, `DaskTaskRunner`, and `RayTaskRunner` to handle concurrent or parallel task execution.

1. *Concurrent Task Execution*:
   Use task runners to execute tasks concurrently. For example, the `ThreadPoolTaskRunner` runs tasks concurrently within a thread pool, allowing you to submit tasks using the `.submit()` method. Each submitted task returns a `PrefectFuture` object, which you can use to access the task's state and result.

   Learn more about task runners for concurrent execution here: <https://docs-3.prefect.io/v3/develop/task-runners|Prefect Task Runners Documentation>

2. *Retrieving Task Results*:
   After submitting tasks using `.submit()`, you can retrieve their results with the `.result()` method on the `PrefectFuture` objects returned. This is useful when you want to wait for tasks to complete and get their output.

   For details on using `.result()`, explore this: <https://docs-3.prefect.io/v3/develop/deferred-tasks|Deferred Tasks Documentation>

By using these features, you can efficiently run tasks concurrently and handle their results without blocking execution. It's like having a team of diligent scribes working in parallel but still getting your important notes together in the end—how utterly delightful!

```In [1]: from prefect import flow, task

In [2]: @flow
   ...: def f(): task(lambda x: x +1).map(range(3)).result()

In [3]: f()
09:24:14.479 | INFO    | prefect.engine - Created flow run 'literate-prawn' for flow 'f'
09:24:14.839 | INFO    | Task run '&lt;lambda&gt;-8d2' - Finished in state Completed()
09:24:14.840 | INFO    | Task run '&lt;lambda&gt;-2cc' - Finished in state Completed()
09:24:14.841 | INFO    | Task run '&lt;lambda&gt;-df1' - Finished in state Completed()
09:24:15.009 | INFO    | Flow run 'literate-prawn' - Finished in state Completed()```

<https://www.youtube.com/watch?v=k74tEYSK_t8>

Hmm. after adding  `.result(raise_on_failure)` after submitting tasks, it seems to only run 1 task at a time

I’m trying to figure out what else could be causing it

basically I have a set of tasks I’m submitting in a for loop. If any of those tasks return an Exception object, I skip the subsequent related task runs in a similar for loop

for any of the first tasks that succeeded, they get submitted for a subsequent task run