[
https://issues.apache.org/jira/browse/IMPALA-10785?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17384646#comment-17384646
]
pengdou1990 edited comment on IMPALA-10785 at 7/21/21, 6:57 AM:
----------------------------------------------------------------
union output tuple layout depence on it's first child node：
[https://github.com/apache/impala/blob/master/fe/src/main/java/org/apache/impala/analysis/SetOperationStmt.java#L599]
each slot's isNullable_ value and null indicator depence on all childnode's
corresponding slot：
[https://github.com/apache/impala/blob/master/fe/src/main/java/org/apache/impala/analysis/SetOperationStmt.java#L644]
in kudu table tuple, the primary key slot‘s isNullable_ = false
in hdfs table tuple, all slot's isNullable_ = true
in kudu table union with hdfs table situation, tuple memory layout may as hdfs
table’s tuple memory layout (string type length is 12 byte， without padding),
and isNullable_ of slots in tuple as kudu table’s
neither hdfs table nor kudu table can pass isChildPassthrough check, so pass
through does not take effect
was (Author: pengdou1990):
union output tuple layout depence on it's first child node：
[https://github.com/apache/impala/blob/master/fe/src/main/java/org/apache/impala/analysis/SetOperationStmt.java#L599]
each slot's isNullable_ value and null indicator depence on all childnode's
corresponding slot：
[https://github.com/apache/impala/blob/master/fe/src/main/java/org/apache/impala/analysis/SetOperationStmt.java#L644]
in kudu table tuple, the primary key slot‘s isNullable_ = false
in hdfs table tuple, all slot's isNullable_ = true
in kudu table union with hdfs table situation, tuple memory layout may as hdfs
table’s tuple memory layout (string type length is 12 byte， without padding),
and isNullable_ of slots in tuple as kudu table’s
neither hdfs table nor kudu table can't pass isChildPassthrough check, so pass
through does not take effect
> when union kudu table and hdfs table, union passthrough does not take effect
> ----------------------------------------------------------------------------
>
> Key: IMPALA-10785
> URL: https://issues.apache.org/jira/browse/IMPALA-10785
> Project: IMPALA
> Issue Type: Improvement
> Reporter: pengdou1990
> Priority: Major
>
> IMPALA-3586 already supports union passthrough, and brings great performance
> improvements in union， but there is still some problems when union between
> hdfs table and kudu table ，several points cause the problem:
> # in kudu scanner node output TupleDescriptor, string slot is 16B,while in
> hdfs scanner node output TupleDescriptor, string slot is 12B,cause tuple
> memory layout mismatch
> # in kudu scanner node output TupleDescriptor, string slot is 16B, while in
> Union output TupleDescriptor, string slot is 12B,cause tuple memory layout
> mismatch
> # in Kudu Scannode, row key slot is not null, while in hdfs node, not null
> slot can't get from the metadata, cause tuple memory layout mismatch
> I hive resolved the 1st and 2nd points, how should I do with the 3rd point?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
