Kontinuation opened a new pull request, #1021:
URL: https://github.com/apache/datafusion-comet/pull/1021

   ## Which issue does this PR close?
   
   This issue relates to https://github.com/apache/datafusion-comet/issues/996 
and https://github.com/apache/datafusion-comet/pull/1004
   
   ## Rationale for this change
   
   Investigating various approaches to simplify memory-related configuration 
and reduce the memory required to run large queries.
   
   ## What changes are included in this PR?
   
   This PR adds a `spark.comet.exec.memoryPool` configuration for easily 
running queries using various memory pool setups.
   
   * `greedy` (default): Each operator has its own GreedyMemoryPool
   * `fair_spill`: Each operator has its own FairSpillPool
   * `fair_spill_task_shared`: All operators for the same task attempt shares 
the same FairSpillPool
   * `greedy_global`: All operators in the same executor instance shares the 
same GreedyMemoryPool
   * `fair_spill_global`: All operators in the same executor instance shares 
the same FairSpillPool
   
   ## How are these changes tested?
   
   This PR is for experimentation purposes and the default configuration has no 
behavioral changes. Other configurations will be tested manually by running 
experiments.
   


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to