Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
Attachments
Attachments
Issue Links
- is related to
-
FLINK-14133 Improve batch sql and hive integrate performance milestone-1
- Closed
1.
|
Introduce parallelism inference for source | Closed | Unassigned |
|
|||||||||
2.
|
Introduce parquet ColumnarRow split reader | Resolved | Jingsong Lee |
|
|||||||||
3.
|
Investigate pushing predicate/projection to underlying Hive input format | Open | Unassigned | ||||||||||
4.
|
Implement LimitableTableSource for FileSystemTableFactory | Closed | Unassigned | ||||||||||
5.
|
Multi vectorized read version support for hive orc read | Closed | Jingsong Lee |
|
|||||||||
6.
|
Provide a new InputSplitAssigner that consider page cache utilization ratio | Open | Unassigned | ||||||||||
7.
|
Reusing record object in StreamTaskNetworkInput | Open | Unassigned | ||||||||||
8.
|
Optimize the operator algorithm to lazily allocate memory | Resolved | Jingsong Lee |
|
|||||||||
9.
|
List/Map/Struct types support for vectorized orc reader | Resolved | Wang Wei | ||||||||||
10.
|
Introduce WritableVectors for abstract writing | Resolved | Jingsong Lee |
|
|||||||||
11.
|
Integrate parquet columnar row reader to hive | Resolved | Jingsong Lee |
|
|||||||||
12.
|
Provide an assigner strategy of average splits allocation | Open | Unassigned |