drin commented on PR #13487:
URL: https://github.com/apache/arrow/pull/13487#issuecomment-1190604497
I wanted to update with initial changes to `hashing_benchmark.cc`. To make
sure comparisons covered at least a few cases, I decided to add FastHash64,
although it is currently very copy pasta.
Barring code style/organization improvements, I was hoping to get some
feedback on whether these benchmarks are fundamentally what we would like to
see. Some questions I have:
- how do I ensure I am not including the time to construct an arrow::Array
from a vector in the benchmark time?
- do we want to time calling the function, or just the mechanisms that the
compute function wraps?
- do we want to do both to also capture any overhead in compute kernel
invocation?
--
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]
