Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

parallelism.png

Hey everyone :slightly_smiling_face:

I have a question about the `pyflyte run --max-parallelism` option of executions.

Even when setting it to 1, I see multiple pods scheduled at the same time while I would have expected only a single one. Am I misunderstanding this option? Is the same for you as well?

What version of Flyte are you running? I know there was a lingering issue where max-parallelism was not correctly tracked on multiple node evaluations within the same propeller round.

You are entirely correct though, we expect only one Pod.

`<http://cr.flyte.org/flyteorg/flytepropeller-release:v1.10.0|cr.flyte.org/flyteorg/flytepropeller-release:v1.10.0>`

We are using a very recent one :thinking_face:

:sweat_smile: interesting. I'll take a look here in a bit.

alright, so IIUC we're cutting a new release today or tomorrow that will <https://github.com/flyteorg/flyte/pull/4318|include this fix>. i just tested again and in 1.11 `max-parallelism` configuration with be respected correctly.

Looking forward to test :slightly_smiling_face:

Ran this in our staging cluster, can confirm it works :slightly_smiling_face: