andygrove opened a new issue #658:
URL: https://github.com/apache/arrow-datafusion/issues/658


   **Describe the bug**
   I ran TPC-H query 12 several times with DataFusion and got different results 
each time:
   
   ```
   +------------+-----------------+----------------+
   | l_shipmode | high_line_count | low_line_count |
   +------------+-----------------+----------------+
   | MAIL       | 623097          | 934685         |
   | SHIP       | 622962          | 934510         |
   +------------+-----------------+----------------+
   Query 12 iteration 0 took 42932.5 ms
   
   +------------+-----------------+----------------+
   | l_shipmode | high_line_count | low_line_count |
   +------------+-----------------+----------------+
   | MAIL       | 623099          | 934685         |
   | SHIP       | 622961          | 934513         |
   +------------+-----------------+----------------+
   Query 12 iteration 0 took 41807.4 ms
   
   +------------+-----------------+----------------+
   | l_shipmode | high_line_count | low_line_count |
   +------------+-----------------+----------------+
   | MAIL       | 623099          | 934691         |
   | SHIP       | 622961          | 934514         |
   +------------+-----------------+----------------+
   Query 12 iteration 1 took 41652.9 ms
   
   +------------+-----------------+----------------+
   | l_shipmode | high_line_count | low_line_count |
   +------------+-----------------+----------------+
   | MAIL       | 623097          | 934686         |
   | SHIP       | 622962          | 934510         |
   +------------+-----------------+----------------+
   Query 12 iteration 2 took 41660.9 ms
   
   ```
   
   I see the same behavior with Ballista.
   
   ```
   Query 12 iteration 0 took 32341.6 ms
   +------------+-----------------+----------------+
   | l_shipmode | high_line_count | low_line_count |
   +------------+-----------------+----------------+
   | MAIL       | 623094          | 934692         |
   | SHIP       | 622961          | 934512         |
   +------------+-----------------+----------------+
   
   Query 12 iteration 1 took 29347.7 ms
   +------------+-----------------+----------------+
   | l_shipmode | high_line_count | low_line_count |
   +------------+-----------------+----------------+
   | MAIL       | 623093          | 934697         |
   | SHIP       | 622963          | 934515         |
   +------------+-----------------+----------------+
   ```
   
   **To Reproduce**
   ```bash
   ../target/release/tpch benchmark datafusion --path 
/mnt/tpch/parquet-sf100-partitioned/ --format parquet --iterations 3 --query 12 
--debug
   ```
   
   **Expected behavior**
   Results should be the same on each run.
   
   **Additional context**
   None
   


-- 
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: github-unsubscr...@arrow.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Reply via email to