lidavidm commented on pull request #9656:
URL: https://github.com/apache/arrow/pull/9656#issuecomment-827188209


   Transferring off of the I/O pool fixes the performance degradation (at some 
cost to the 16-file S3 case, but even then, we're still much faster than 
before). This also gets us in about the same position as we were when testing 
ARROW-7001.
   
   ![Local Median Scan Time 
(seconds)(2)](https://user-images.githubusercontent.com/327919/116159844-f76c5080-a6be-11eb-8bfe-8ce1f6df3b8f.png)
   ![S3 Median Scan Time 
(seconds)(6)](https://user-images.githubusercontent.com/327919/116159849-f9361400-a6be-11eb-9f17-7943f006c298.png)


-- 
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.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to