etseidl commented on pull request #11131:
URL: https://github.com/apache/arrow/pull/11131#issuecomment-918406821


   Ok, I modified the stream_reader_writer example to just do the write, and 
changed the number of rows from 2000 to 20M (and the max row group size from 
1000 to 10M).  I ran this on my laptop with and without the changes in this 
patch.  I used the csh time command to compare run times. The results are:
   ```
   original:
   52.919u 2.146s 0:55.81 98.6% 0+0k 0+2646928io 0pf+0w
   53.328u 2.327s 0:56.38 98.6% 0+0k 0+2646928io 0pf+0w
   53.048u 2.341s 0:56.06 98.7% 0+0k 0+2646928io 0pf+0w
   
   patched:
   35.367u 2.012s 0:37.88 98.6% 0+0k 0+2646928io 0pf+0w
   35.133u 1.717s 0:37.36 98.6% 0+0k 0+2646928io 0pf+0w
   35.228u 1.950s 0:37.61 98.8% 0+0k 0+2646928io 0pf+0w
   ```
   @emkornfield Do you want benchmark code added to the PR, or are these 
results sufficient?


-- 
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