Github user cloud-fan commented on the pull request:
https://github.com/apache/spark/pull/12067#issuecomment-207195178
In the benchmark, for RDD, we first apply a function to turn a long into a
`Data`, then do aggregate. For Dataset, we first turn a long to a `UTFString`,
then turn the `UTF8String` to `String` and create a `Data`, then do aggregate.
This may be one reason why Dataset is slower than RDD.
---
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]
