GitHub user kunwp1 added a comment to the discussion: Proposal: Automatically terminate idle Kubernetes Computing Units
I think the proposed design is a great starting point. One concern I have is around long-running workflows: users will likely trigger a run and return later to check the results. If they return after a timeout period and the CU has been garbage collected, they won't be able to see the execution results. It might be worth considering this scenario. GitHub link: https://github.com/apache/texera/discussions/6264#discussioncomment-17624089 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
