flyteorg / flyte

Scalable and flexible workflow orchestration platform that seamlessly unifies data, ML and analytics stacks.
https://flyte.org
Apache License 2.0
5.47k stars 584 forks source link

[UI Feature] Task node overrides are not reflected in flyteconsole #5589

Open eapolinario opened 2 months ago

eapolinario commented 2 months ago

Motivation: Why do you think this is important?

The Task Details tab in flyteconsole is not aware of resource overridden in a task, e.g.:

from flytekit import task, Resources, workflow

@task(
    requests=Resources(
        cpu="2",
        mem="2Gi",
    ),
)
def foo():
    print('task')

@workflow
def my_wf():
    foo()
    foo().with_overrides(
        requests=Resources(
            cpu="1",
            mem="1Gi",
        ),
    )

Notice how in the second execution of foo we override requests.

Currently the Task details view for the second task displays the original requests value.

Goal: What should the final outcome look like, ideally?

The Task Details view should be aware of resource overrides so as not to confuse users, since not having them there gives the false impression that those are not being applied, when in reality they are.

Propose: Link/Inline OR Additional context

The Task Details view for a given task execution should introspect the workflow closure to find which resources are being overridden (if any) and apply that transformation before rendering them.

Are you sure this issue hasn't been raised already?

Have you read the Code of Conduct?

runllm[bot] commented 2 months ago

Based on Flyte Issue #4543, the Task Details tab in flyteconsole does not currently reflect resource overrides applied using with_overrides. This can lead to confusion as users might think the overrides are not being applied when they actually are.

Would you like more detail on how to address this issue or further explanation on the current behavior and potential fixes?

Sources:


This is an AI-generated response and your feedback is appreciated! Please leave a 👍 if this is helpful and 👎 if it is not.

Mention @runllm to ask follow-up questions. See runllm.com for more!