The cloud-native orchestrator for the whole development lifecycle, with integrated lineage and observability.

dagster

Hi everyone. I can use `dagster-k8s/config` in `tags` when using `define_asset_job` as a way to define what cpu and memory my jobs will have in my kubernetes deployment. However, I find that when I refresh assets manually, I tend to want to materialize an individual asset or an asset group outside of the job context in Dagit. When I do this, the default memory and cpu is used. Is it possible to use `dagster-k8s/config` when defining an asset or asset group?

It looks like I can go into a job, filter to an asset group, and trigger from there. That will use my `dagster-k8s/config` :dagster-spin:

You can also set a default for the whole helm chart here: <https://github.com/dagster-io/dagster/blob/master/helm/dagster/values.yaml#L526-L537>