alamb commented on PR #9595:
URL:
https://github.com/apache/arrow-datafusion/pull/9595#issuecomment-2012974133
Here is the result of running the entire benchmark:
```
group dfschema
main
----- --------
----
logical_aggregate_with_join 1.13 1366.0±39.42µs ?
?/sec 1.00 1210.0±10.10µs ? ?/sec
logical_plan_tpch_all 1.00 16.4±0.16ms ?
?/sec 1.00 16.3±0.13ms ? ?/sec
logical_select_all_from_1000 1.00 25.7±0.18ms ?
?/sec 3.86 98.9±0.42ms ? ?/sec
logical_select_one_from_700 1.20 837.5±10.22µs ?
?/sec 1.00 696.2±17.76µs ? ?/sec
logical_trivial_join_high_numbered_columns 1.04 787.7±7.54µs ?
?/sec 1.00 754.7±9.10µs ? ?/sec
logical_trivial_join_low_numbered_columns 1.07 768.8±10.41µs ?
?/sec 1.00 719.7±8.67µs ? ?/sec
physical_plan_tpch_all 1.00 127.1±0.50ms ?
?/sec 1.01 128.8±0.74ms ? ?/sec
physical_plan_tpch_q1 1.00 7.5±0.04ms ?
?/sec 1.03 7.7±0.05ms ? ?/sec
physical_plan_tpch_q10 1.00 6.1±0.05ms ?
?/sec 1.02 6.2±0.03ms ? ?/sec
physical_plan_tpch_q11 1.04 5.0±0.02ms ?
?/sec 1.00 4.8±0.03ms ? ?/sec
physical_plan_tpch_q12 1.02 3.9±0.02ms ?
?/sec 1.00 3.9±0.02ms ? ?/sec
physical_plan_tpch_q13 1.02 2.6±0.01ms ?
?/sec 1.00 2.6±0.02ms ? ?/sec
physical_plan_tpch_q14 1.03 3.4±0.03ms ?
?/sec 1.00 3.3±0.02ms ? ?/sec
physical_plan_tpch_q16 1.01 5.0±0.02ms ?
?/sec 1.00 5.0±0.02ms ? ?/sec
physical_plan_tpch_q17 1.05 4.9±0.02ms ?
?/sec 1.00 4.6±0.03ms ? ?/sec
physical_plan_tpch_q18 1.03 5.4±0.03ms ?
?/sec 1.00 5.3±0.03ms ? ?/sec
physical_plan_tpch_q19 1.01 9.6±0.04ms ?
?/sec 1.00 9.5±0.05ms ? ?/sec
physical_plan_tpch_q2 1.00 11.5±0.04ms ?
?/sec 1.03 11.9±0.08ms ? ?/sec
physical_plan_tpch_q20 1.04 6.4±0.05ms ?
?/sec 1.00 6.1±0.04ms ? ?/sec
physical_plan_tpch_q21 1.01 9.2±0.05ms ?
?/sec 1.00 9.1±0.05ms ? ?/sec
physical_plan_tpch_q22 1.01 4.4±0.02ms ?
?/sec 1.00 4.4±0.02ms ? ?/sec
physical_plan_tpch_q3 1.02 4.1±0.02ms ?
?/sec 1.00 4.0±0.02ms ? ?/sec
physical_plan_tpch_q4 1.00 3.0±0.02ms ?
?/sec 1.08 3.2±0.01ms ? ?/sec
physical_plan_tpch_q5 1.06 6.2±0.20ms ?
?/sec 1.00 5.9±0.06ms ? ?/sec
physical_plan_tpch_q6 1.02 1977.9±12.03µs ?
?/sec 1.00 1930.7±10.86µs ? ?/sec
physical_plan_tpch_q7 1.00 8.1±0.03ms ?
?/sec 1.02 8.3±0.04ms ? ?/sec
physical_plan_tpch_q8 1.00 10.6±0.06ms ?
?/sec 1.11 11.8±0.06ms ? ?/sec
physical_plan_tpch_q9 1.00 8.0±0.04ms ?
?/sec 1.13 9.0±0.05ms ? ?/sec
physical_select_all_from_1000 1.00 175.0±0.52ms ?
?/sec 4.14 724.3±1.39ms ? ?/sec
physical_select_one_from_700 1.23 4.9±0.02ms ?
?/sec 1.00 4.0±0.02ms ? ?/sec
alamb@aal-dev:~/arrow-datafusion4$
```
The queries that are reported slower are all very low latency to begin with
and I think may be more measurement noise
I think the "all" bechmarks are more meaningful as they do more work / more
queries
--
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]
