[
https://issues.apache.org/jira/browse/HIVE-8842?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Szehon Ho updated HIVE-8842:
----------------------------
Description:
Enabling the SparkMapJoinResolver and SparkReduceSinkMapJoinProc, I see the
following:
{noformat}
explain select * from src src1 JOIN src src2 ON (src1.key = src2.key) JOIN src
src3 ON (src1.key + src2.key = src3.key);
{noformat}
produces too many stages (six), and too many HashTableSink.
{noformat}
STAGE DEPENDENCIES:
Stage-5 is a root stage
Stage-4 depends on stages: Stage-5
Stage-3 depends on stages: Stage-4
Stage-7 is a root stage
Stage-6 depends on stages: Stage-7
Stage-0 is a root stage
STAGE PLANS:
Stage: Stage-5
Spark
DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:3
Vertices:
Map 1
Map Operator Tree:
TableScan
alias: src2
Statistics: Num rows: 29 Data size: 5812 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats:
COMPLETE Column stats: NONE
HashTable Sink Operator
condition expressions:
0 {key} {value}
1 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
Stage: Stage-4
Spark
DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:2
Vertices:
Map 3
Map Operator Tree:
TableScan
alias: src1
Statistics: Num rows: 29 Data size: 5812 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats:
COMPLETE Column stats: NONE
Map Join Operator
condition map:
Inner Join 0 to 1
condition expressions:
0 {key} {value}
1 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
outputColumnNames: _col0, _col1, _col5, _col6
input vertices:
1 Map 1
Statistics: Num rows: 16 Data size: 3306 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: (_col0 + _col5) is not null (type: boolean)
Statistics: Num rows: 8 Data size: 1653 Basic stats:
COMPLETE Column stats: NONE
HashTable Sink Operator
condition expressions:
0 {_col0} {_col1} {_col5} {_col6}
1 {key} {value}
keys:
0 (_col0 + _col5) (type: double)
1 UDFToDouble(key) (type: double)
Stage: Stage-3
Spark
DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:1
Vertices:
Map 2
Map Operator Tree:
TableScan
alias: src3
Statistics: Num rows: 29 Data size: 5812 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: UDFToDouble(key) is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats:
COMPLETE Column stats: NONE
Map Join Operator
condition map:
Inner Join 0 to 1
condition expressions:
0 {_col0} {_col1} {_col5} {_col6}
1 {key} {value}
keys:
0 (_col0 + _col5) (type: double)
1 UDFToDouble(key) (type: double)
outputColumnNames: _col0, _col1, _col5, _col6, _col10,
_col11
input vertices:
0 Map 3
Statistics: Num rows: 16 Data size: 3306 Basic stats:
COMPLETE Column stats: NONE
Select Operator
expressions: _col0 (type: string), _col1 (type:
string), _col5 (type: string), _col6 (type: string), _col10 (type: string),
_col11 (type: string)
outputColumnNames: _col0, _col1, _col2, _col3, _col4,
_col5
Statistics: Num rows: 16 Data size: 3306 Basic stats:
COMPLETE Column stats: NONE
File Output Operator
compressed: false
Statistics: Num rows: 16 Data size: 3306 Basic stats:
COMPLETE Column stats: NONE
table:
input format:
org.apache.hadoop.mapred.TextInputFormat
output format:
org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
serde:
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
Stage: Stage-7
Spark
DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:3
Vertices:
Map 1
Map Operator Tree:
TableScan
alias: src2
Statistics: Num rows: 29 Data size: 5812 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats:
COMPLETE Column stats: NONE
HashTable Sink Operator
condition expressions:
0 {key} {value}
1 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
Stage: Stage-6
Spark
DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:2
Vertices:
Map 3
Map Operator Tree:
TableScan
alias: src1
Statistics: Num rows: 29 Data size: 5812 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats:
COMPLETE Column stats: NONE
Map Join Operator
condition map:
Inner Join 0 to 1
condition expressions:
0 {key} {value}
1 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
outputColumnNames: _col0, _col1, _col5, _col6
input vertices:
1 Map 1
Statistics: Num rows: 16 Data size: 3306 Basic stats:
COMPLETE Column stats: NONE
Filter Operator
predicate: (_col0 + _col5) is not null (type: boolean)
Statistics: Num rows: 8 Data size: 1653 Basic stats:
COMPLETE Column stats: NONE
HashTable Sink Operator
condition expressions:
0 {_col0} {_col1} {_col5} {_col6}
1 {key} {value}
keys:
0 (_col0 + _col5) (type: double)
1 UDFToDouble(key) (type: double)
Stage: Stage-0
Fetch Operator
limit: -1
Processor Tree:
ListSink
{noformat}
was:
Enabling the SparkMapJoinResolver and SparkReduceSinkMapJoinProc, I see the
following:
{noformat}
explain select * from src src1 JOIN src src2 ON (src1.key = src2.key) JOIN src
src3 ON (src1.key + src2.key = src3.key);
{noformat}
produces too many stages (six), and too many HashTableSink.
{noformat}
STAGE DEPENDENCIES:
Stage-5 is a root stage
Stage-4 depends on stages: Stage-5
Stage-0 depends on stages: Stage-4
STAGE PLANS:
Stage: Stage-5
Map Reduce Local Work
Alias -> Map Local Tables:
src1
Fetch Operator
limit: -1
src2
Fetch Operator
limit: -1
Alias -> Map Local Operator Tree:
src1
TableScan
alias: src1
Statistics: Num rows: 29 Data size: 5812 Basic stats: COMPLETE
Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats: COMPLETE
Column stats: NONE
HashTable Sink Operator
condition expressions:
0 {value}
1 {value}
2 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
2 key (type: string)
src2
TableScan
alias: src2
Statistics: Num rows: 29 Data size: 5812 Basic stats: COMPLETE
Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats: COMPLETE
Column stats: NONE
HashTable Sink Operator
condition expressions:
0 {value}
1 {value}
2 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
2 key (type: string)
Stage: Stage-4
Map Reduce
Map Operator Tree:
TableScan
alias: src3
Statistics: Num rows: 29 Data size: 5812 Basic stats: COMPLETE
Column stats: NONE
Filter Operator
predicate: key is not null (type: boolean)
Statistics: Num rows: 15 Data size: 3006 Basic stats: COMPLETE
Column stats: NONE
Map Join Operator
condition map:
Inner Join 0 to 1
Inner Join 1 to 2
condition expressions:
0 {key} {value}
1 {key} {value}
2 {key} {value}
keys:
0 key (type: string)
1 key (type: string)
2 key (type: string)
outputColumnNames: _col0, _col1, _col5, _col6, _col10, _col11
Statistics: Num rows: 33 Data size: 6613 Basic stats: COMPLETE
Column stats: NONE
Select Operator
expressions: _col0 (type: string), _col1 (type: string),
_col5 (type: string), _col6 (type: string), _col10 (type: string), _col11
(type: string)
outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5
Statistics: Num rows: 33 Data size: 6613 Basic stats:
COMPLETE Column stats: NONE
File Output Operator
compressed: false
Statistics: Num rows: 33 Data size: 6613 Basic stats:
COMPLETE Column stats: NONE
table:
input format: org.apache.hadoop.mapred.TextInputFormat
output format:
org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
serde:
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
Local Work:
Map Reduce Local Work
Stage: Stage-0
Fetch Operator
limit: -1
Processor Tree:
ListSink
{noformat}
> auto_join2.q produces incorrect tree [Spark Branch]
> ---------------------------------------------------
>
> Key: HIVE-8842
> URL: https://issues.apache.org/jira/browse/HIVE-8842
> Project: Hive
> Issue Type: Sub-task
> Components: Spark
> Reporter: Szehon Ho
>
> Enabling the SparkMapJoinResolver and SparkReduceSinkMapJoinProc, I see the
> following:
> {noformat}
> explain select * from src src1 JOIN src src2 ON (src1.key = src2.key) JOIN
> src src3 ON (src1.key + src2.key = src3.key);
> {noformat}
> produces too many stages (six), and too many HashTableSink.
> {noformat}
> STAGE DEPENDENCIES:
> Stage-5 is a root stage
> Stage-4 depends on stages: Stage-5
> Stage-3 depends on stages: Stage-4
> Stage-7 is a root stage
> Stage-6 depends on stages: Stage-7
> Stage-0 is a root stage
> STAGE PLANS:
> Stage: Stage-5
> Spark
> DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:3
> Vertices:
> Map 1
> Map Operator Tree:
> TableScan
> alias: src2
> Statistics: Num rows: 29 Data size: 5812 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: key is not null (type: boolean)
> Statistics: Num rows: 15 Data size: 3006 Basic stats:
> COMPLETE Column stats: NONE
> HashTable Sink Operator
> condition expressions:
> 0 {key} {value}
> 1 {key} {value}
> keys:
> 0 key (type: string)
> 1 key (type: string)
> Stage: Stage-4
> Spark
> DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:2
> Vertices:
> Map 3
> Map Operator Tree:
> TableScan
> alias: src1
> Statistics: Num rows: 29 Data size: 5812 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: key is not null (type: boolean)
> Statistics: Num rows: 15 Data size: 3006 Basic stats:
> COMPLETE Column stats: NONE
> Map Join Operator
> condition map:
> Inner Join 0 to 1
> condition expressions:
> 0 {key} {value}
> 1 {key} {value}
> keys:
> 0 key (type: string)
> 1 key (type: string)
> outputColumnNames: _col0, _col1, _col5, _col6
> input vertices:
> 1 Map 1
> Statistics: Num rows: 16 Data size: 3306 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: (_col0 + _col5) is not null (type: boolean)
> Statistics: Num rows: 8 Data size: 1653 Basic stats:
> COMPLETE Column stats: NONE
> HashTable Sink Operator
> condition expressions:
> 0 {_col0} {_col1} {_col5} {_col6}
> 1 {key} {value}
> keys:
> 0 (_col0 + _col5) (type: double)
> 1 UDFToDouble(key) (type: double)
> Stage: Stage-3
> Spark
> DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:1
> Vertices:
> Map 2
> Map Operator Tree:
> TableScan
> alias: src3
> Statistics: Num rows: 29 Data size: 5812 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: UDFToDouble(key) is not null (type: boolean)
> Statistics: Num rows: 15 Data size: 3006 Basic stats:
> COMPLETE Column stats: NONE
> Map Join Operator
> condition map:
> Inner Join 0 to 1
> condition expressions:
> 0 {_col0} {_col1} {_col5} {_col6}
> 1 {key} {value}
> keys:
> 0 (_col0 + _col5) (type: double)
> 1 UDFToDouble(key) (type: double)
> outputColumnNames: _col0, _col1, _col5, _col6, _col10,
> _col11
> input vertices:
> 0 Map 3
> Statistics: Num rows: 16 Data size: 3306 Basic stats:
> COMPLETE Column stats: NONE
> Select Operator
> expressions: _col0 (type: string), _col1 (type:
> string), _col5 (type: string), _col6 (type: string), _col10 (type: string),
> _col11 (type: string)
> outputColumnNames: _col0, _col1, _col2, _col3, _col4,
> _col5
> Statistics: Num rows: 16 Data size: 3306 Basic stats:
> COMPLETE Column stats: NONE
> File Output Operator
> compressed: false
> Statistics: Num rows: 16 Data size: 3306 Basic
> stats: COMPLETE Column stats: NONE
> table:
> input format:
> org.apache.hadoop.mapred.TextInputFormat
> output format:
> org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
> serde:
> org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
> Stage: Stage-7
> Spark
> DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:3
> Vertices:
> Map 1
> Map Operator Tree:
> TableScan
> alias: src2
> Statistics: Num rows: 29 Data size: 5812 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: key is not null (type: boolean)
> Statistics: Num rows: 15 Data size: 3006 Basic stats:
> COMPLETE Column stats: NONE
> HashTable Sink Operator
> condition expressions:
> 0 {key} {value}
> 1 {key} {value}
> keys:
> 0 key (type: string)
> 1 key (type: string)
> Stage: Stage-6
> Spark
> DagName: szehon_20141112105656_dd50e07d-94ad-4f9d-899e-bcb6d9a39c13:2
> Vertices:
> Map 3
> Map Operator Tree:
> TableScan
> alias: src1
> Statistics: Num rows: 29 Data size: 5812 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: key is not null (type: boolean)
> Statistics: Num rows: 15 Data size: 3006 Basic stats:
> COMPLETE Column stats: NONE
> Map Join Operator
> condition map:
> Inner Join 0 to 1
> condition expressions:
> 0 {key} {value}
> 1 {key} {value}
> keys:
> 0 key (type: string)
> 1 key (type: string)
> outputColumnNames: _col0, _col1, _col5, _col6
> input vertices:
> 1 Map 1
> Statistics: Num rows: 16 Data size: 3306 Basic stats:
> COMPLETE Column stats: NONE
> Filter Operator
> predicate: (_col0 + _col5) is not null (type: boolean)
> Statistics: Num rows: 8 Data size: 1653 Basic stats:
> COMPLETE Column stats: NONE
> HashTable Sink Operator
> condition expressions:
> 0 {_col0} {_col1} {_col5} {_col6}
> 1 {key} {value}
> keys:
> 0 (_col0 + _col5) (type: double)
> 1 UDFToDouble(key) (type: double)
> Stage: Stage-0
> Fetch Operator
> limit: -1
> Processor Tree:
> ListSink
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
