Github user kanzhang commented on the pull request:
https://github.com/apache/spark/pull/829#issuecomment-44786769
How about we keep the row format behavior and add columnar format behavior
separately? For example, ```SchemaRDD.cache()``` and ```unpersist()``` stay as
is and they operate on the row format. Independently, we add
SchemaRDD.registerAsTable(tableName) and SchemaRDD.cacheAsTable(tableName),
which operates on the columnar format. The latter will automatically register
it as table and cache the table as ```InMemoryColumnarTableScan```. It can
happen that a SchemaRDD is cached in both row and columnar formats, if the user
so chooses.
---
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.
---
