parthchandra commented on PR #1034:
URL:
https://github.com/apache/datafusion-comet/pull/1034#issuecomment-2460421886
@andygrove I tried with your change and I get -
```
Running benchmark: ColumnarToRowExec
Running case: Spark Columnar To Row - integer
Stopped after 35 iterations, 2037 ms
Running case: Comet Columnar To Row - integer
Stopped after 15 iterations, 2022 ms
OpenJDK 64-Bit Server VM 11.0.19+7-LTS on Mac OS X 14.6
Apple M3 Max
ColumnarToRowExec: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
Spark Columnar To Row - integer 42 58
13 249.5 4.0 1.0X
Comet Columnar To Row - integer 52 135
36 200.9 5.0 0.8X
```
So not really any better. I also tried with another change so that I
precompute row lengths and rows start addresses to accomodate variable length
fields and look up the row lengths as I iterate over the rows for each column.
That slowed down to `0.7x` of Spark.
I don't think this is going to work, so I'm leaving this closed.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
