Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Issue #736] pipeline execution of hash partitioned join (cont.) #771

Open
wants to merge 63 commits into
base: master
Choose a base branch
from

Conversation

jasha64
Copy link
Contributor

@jasha64 jasha64 commented Nov 6, 2024

  1. Fix support for post partition in partitioned join stream workers.
  2. Fix various hardcode in partition stream workers.
  3. Fix bug with empty partition results: If nation and region are in the join query, some of their partitions will be empty. In legacy non-streaming mode, the next-level workers can simply write nothing to the intermediate file, and skip the join if the joiner hash table is empty. But in streaming mode, we still have to send an empty row group to next-level workers, and still have to listen on joiner ports even if its hash table is empty, to enforce the streaming protocol - otherwise, we cannot distinguish between empty intermediate data and worker failure.
  4. Fix a glitch in VhiveInvoker, which could cause the trino startup process to get stuck. Its constructor should not be blocking.

…end streaming support to post partition of joins
…end streaming support to post partition of joins
…ing storage infos in PixelsPlanner"

This reverts commit 09bc92c.
@bianhq bianhq requested review from bianhq and huasiy November 8, 2024 01:57
jasha64 and others added 30 commits December 6, 2024 18:33
…end streaming support to post partition of joins
…end streaming support to post partition of joins
…ing storage infos in PixelsPlanner"

This reverts commit 09bc92c.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants