[
https://issues.apache.org/jira/browse/FLINK-12563?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jingsong Lee updated FLINK-12563:
---------------------------------
Description:
Introduce vector data format to high performance parquet/orc source.
Introduce VectorizedColumnBatch: A VectorizedColumnBatch is a set of rows,
organized with each column as a vector.
Introduce ColumnVector with subclasses: IntColumnVector, LongColumnVector and
etc..
Introduce ColumnarRow: wrap VectorizedColumnBatch as a BaseRow.
was:
Introduce VectorizedColumnBatch: A VectorizedColumnBatch is a set of rows,
organized with each column as a vector.
Introduce ColumnVector with subclasses: IntColumnVector, LongColumnVector and
etc..
Introduce ColumnarRow: wrap VectorizedColumnBatch as a BaseRow.
> Introduce vector data format in blink
> -------------------------------------
>
> Key: FLINK-12563
> URL: https://issues.apache.org/jira/browse/FLINK-12563
> Project: Flink
> Issue Type: Improvement
> Components: Table SQL / Runtime
> Reporter: Jingsong Lee
> Assignee: Jingsong Lee
> Priority: Major
>
> Introduce vector data format to high performance parquet/orc source.
> Introduce VectorizedColumnBatch: A VectorizedColumnBatch is a set of rows,
> organized with each column as a vector.
> Introduce ColumnVector with subclasses: IntColumnVector, LongColumnVector and
> etc..
> Introduce ColumnarRow: wrap VectorizedColumnBatch as a BaseRow.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
