korowa opened a new pull request, #5632:
URL: https://github.com/apache/arrow-datafusion/pull/5632

   # Which issue does this PR close?
   
   <!--
   We generally require a GitHub issue to be filed for all bug fixes and 
enhancements and this helps us generate change logs for our releases. You can 
link an issue to this PR using the GitHub syntax. For example `Closes #123` 
indicates that this PR will close issue #123.
   -->
   
   Closes #5220.
   
   # Rationale for this change
   
   <!--
    Why are you proposing this change? If this is already explained clearly in 
the issue then this section is not needed.
    Explaining clearly why changes are proposed helps reviewers understand your 
changes and offer better suggestions for fixes.  
   -->
   
   Control over memory allocation in `SortMergeJoinExec`
   
   # What changes are included in this PR?
   
   <!--
   There is no need to duplicate the description in the issue here but it is 
sometimes worth providing a summary of the individual changes in this PR.
   -->
   
   - `SMJStream.reservation` -- memory reservation to control buffered data 
size -- it may require significant amount of memory, for example, in case of 
multiple batches with the same join key values
   - `peak_memory_used` -- during join execution, normally, both `try_grow` and 
`shrink` methods are called  -- on adding new buffered batches and on flushing 
them if join key has changed, respectively. In this case it seems reasonable to 
track only peak memory for each partition -- after that it's summed up across 
all partitions which is not really precise, but still valuable in terms of 
showing "worst case" (in fact peak memory used, probably, can be lower)
   - memory limit tests now are able to run with configurable `SessionConfig` 
and perform multiple checks on returned error (used to ensure that memory 
allocation failed in specific operator/stream)
   
   # Are these changes tested?
   
   <!--
   We typically require tests for all PRs in order to:
   1. Prevent the code from being accidentally broken by subsequent changes
   2. Serve as another way to document the expected behavior of the code
   
   If tests are not included in your PR, please explain why (for example, are 
they covered by existing tests)?
   -->
   
   - existsing test cases (shrinks don't lead to negative size)
   - overallocation tests for all join types (if accepted by merge join 
operator)
   - SQL-level test for overallocation
   
   # Are there any user-facing changes?
   
   <!--
   If there are user-facing changes then we may require documentation to be 
updated before approving the PR.
   -->
   
   `SortMergeJoinExec` should now respect runtime memory limitations.
   
   <!--
   If there are any breaking changes to public APIs, please add the `api 
change` label.
   -->


-- 
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]

Reply via email to