Github user leifwalsh commented on the issue:
https://github.com/apache/spark/pull/15821
The next iteration of this for perf would likely involve generating the
arrow batches on executors and having the driver use the new streaming arrow
format to just forward this to python. In our experiments, assembling arrays of
internal rows dominates time, then transposing them and forming an arrow record
batch is pretty quick. If we can do that work in parallel on the executors,
we're likely to get another big win.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
