Gaurav598 commented on PR #40496: URL: https://github.com/apache/beam/pull/40496#issuecomment-6093629097
R: @tvalentyn Hi @tvalentyn, I reviewed your feedback on #40283, particularly the concerns around nested Dask computations, potential worker deadlocks, and changing the default behavior. I've updated this PR to make partition-at-a-time evaluation opt-in, preserving the existing default behavior. Both paths now use `worker_client()` for nested computations inside Dask workers, and I've added distributed regression tests covering single-worker/single-thread and multi-worker execution. The focused and distributed tests pass locally. The PR description also documents the remaining memory and performance trade-offs. Would you be willing to review the updated approach when you have time? I'd appreciate your feedback. Thanks! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
