Dandandan commented on pull request #792:
URL: https://github.com/apache/arrow-datafusion/pull/792#issuecomment-890547319
> The implementation is almost finished. I plan to add more tests and make
all cases pass next.
>
> I tried to run the tpc-h benchmark and got a bad result 😕
>
> with cse:
>
> > Running benchmarks with the following options: DataFusionBenchmarkOpt {
query: 1, debug: false, iterations: 5, concurrency: 2, batch_size: 4096, path:
"./data", file_format: "tbl", mem_table: false, partitions: 8 }
> > Query 1 iteration 0 took 2901.0 ms
> > Query 1 iteration 1 took 2767.9 ms
> > Query 1 iteration 2 took 3011.2 ms
> > Query 1 iteration 3 took 2861.0 ms
> > Query 1 iteration 4 took 2955.2 ms
> > Query 1 avg time: 2899.27 ms
>
> master:
>
> > Running benchmarks with the following options: DataFusionBenchmarkOpt {
query: 1, debug: false, iterations: 5, concurrency: 2, batch_size: 4096, path:
"./data", file_format: "tbl", mem_table: false, partitions: 8 }
> > Query 1 iteration 0 took 2751.7 ms
> > Query 1 iteration 1 took 2703.5 ms
> > Query 1 iteration 2 took 2787.9 ms
> > Query 1 iteration 3 took 2863.1 ms
> > Query 1 iteration 4 took 2831.3 ms
> > Query 1 avg time: 2787.50 ms
Could be within the noise. Maybe you can check / log whether the
optimization is applied at all and for which query?
--
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]
