ihji commented on a change in pull request #12786:
URL: https://github.com/apache/beam/pull/12786#discussion_r485383858
##########
File path:
sdks/java/io/parquet/src/main/java/org/apache/beam/sdk/io/parquet/ParquetIO.java
##########
@@ -135,6 +135,16 @@
* ...
* }</pre>
*
+ * <p>Reading with projection can be enabled with the projection schema as
following. The
+ * projection_schema contains only the column that we would like to read and
encoder_schema contains
+ * all field but with the unwanted columns changed to nullable.
+ *
Review comment:
The description is still a bit vague. Can you please elaborate more on
how much improvement users could expect from the column projection? For
example, if the transform with the column projection only reads a half of the
columns, does it also use only half of the memory and finish two times faster
than the transform without the column projection? (If not, why is that?)
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
