[
https://issues.apache.org/jira/browse/HIVE-15357?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15745110#comment-15745110
]
Rui Li commented on HIVE-15357:
-------------------------------
When I tried to re-enable these tests, I found this issue with DPP.
Running the following query in our qtest:
{code}
EXPLAIN select count(*) from srcpart join (select ds as ds, ds as `date` from
srcpart group by ds) s on (srcpart.ds = s.ds) where s.`date` = '2008-04-08';
{code}
will get this following plan:
{noformat}
STAGE DEPENDENCIES:
Stage-2 is a root stage
Stage-1 depends on stages: Stage-2
Stage-0 depends on stages: Stage-1
STAGE PLANS:
Stage: Stage-2
Spark
Edges:
Reducer 7 <- Map 6 (GROUP, 1)
DagName: lirui_20161213205144_16c810ef-52f6-40c9-a346-7dbd2c9ef99e:2
Vertices:
Map 6
Map Operator Tree:
TableScan
alias: srcpart
Statistics: Num rows: 1000 Data size: 10624 Basic stats:
COMPLETE Column stats: NONE
Select Operator
Statistics: Num rows: 1000 Data size: 10624 Basic stats:
COMPLETE Column stats: NONE
Group By Operator
keys: '2008-04-08' (type: string)
mode: hash
outputColumnNames: _col0
Statistics: Num rows: 1000 Data size: 10624 Basic stats:
COMPLETE Column stats: NONE
Reduce Output Operator
key expressions: '2008-04-08' (type: string)
sort order: +
Map-reduce partition columns: '2008-04-08' (type:
string)
Statistics: Num rows: 1000 Data size: 10624 Basic
stats: COMPLETE Column stats: NONE
Map 8
Map Operator Tree:
TableScan
alias: srcpart
Statistics: Num rows: 2000 Data size: 21248 Basic stats:
COMPLETE Column stats: NONE
Select Operator
expressions: ds (type: string)
outputColumnNames: _col0
Statistics: Num rows: 2000 Data size: 21248 Basic stats:
COMPLETE Column stats: NONE
Group By Operator
keys: _col0 (type: string)
mode: hash
outputColumnNames: _col0
Statistics: Num rows: 2000 Data size: 21248 Basic stats:
COMPLETE Column stats: NONE
Spark Partition Pruning Sink Operator
partition key expr: ds
Statistics: Num rows: 2000 Data size: 21248 Basic
stats: COMPLETE Column stats: NONE
target column name: ds
target work: Map 1
Reducer 7
Reduce Operator Tree:
Group By Operator
keys: '2008-04-08' (type: string)
mode: mergepartial
outputColumnNames: _col0
Statistics: Num rows: 500 Data size: 5312 Basic stats: COMPLETE
Column stats: NONE
Select Operator
Statistics: Num rows: 500 Data size: 5312 Basic stats:
COMPLETE Column stats: NONE
Select Operator
expressions: '2008-04-08' (type: string)
outputColumnNames: _col0
Statistics: Num rows: 500 Data size: 5312 Basic stats:
COMPLETE Column stats: NONE
Group By Operator
keys: _col0 (type: string)
mode: hash
outputColumnNames: _col0
Statistics: Num rows: 500 Data size: 5312 Basic stats:
COMPLETE Column stats: NONE
Spark Partition Pruning Sink Operator
partition key expr: ds
Statistics: Num rows: 500 Data size: 5312 Basic stats:
COMPLETE Column stats: NONE
target column name: ds
target work: Map 5
Stage: Stage-1
Spark
Edges:
Reducer 2 <- Map 1 (GROUP, 1)
Reducer 3 <- Map 5 (PARTITION-LEVEL SORT, 1), Reducer 2
(PARTITION-LEVEL SORT, 1)
Reducer 4 <- Reducer 3 (GROUP, 1)
DagName: lirui_20161213205144_16c810ef-52f6-40c9-a346-7dbd2c9ef99e:1
Vertices:
Map 1
Map Operator Tree:
TableScan
alias: srcpart
Statistics: Num rows: 1000 Data size: 10624 Basic stats:
COMPLETE Column stats: NONE
Select Operator
Statistics: Num rows: 1000 Data size: 10624 Basic stats:
COMPLETE Column stats: NONE
Group By Operator
keys: '2008-04-08' (type: string)
mode: hash
outputColumnNames: _col0
Statistics: Num rows: 1000 Data size: 10624 Basic stats:
COMPLETE Column stats: NONE
Reduce Output Operator
key expressions: '2008-04-08' (type: string)
sort order: +
Map-reduce partition columns: '2008-04-08' (type:
string)
Statistics: Num rows: 1000 Data size: 10624 Basic
stats: COMPLETE Column stats: NONE
Map 5
Map Operator Tree:
TableScan
alias: srcpart
Statistics: Num rows: 2000 Data size: 21248 Basic stats:
COMPLETE Column stats: NONE
Reduce Output Operator
key expressions: ds (type: string)
sort order: +
Map-reduce partition columns: ds (type: string)
Statistics: Num rows: 2000 Data size: 21248 Basic stats:
COMPLETE Column stats: NONE
Reducer 2
Reduce Operator Tree:
Group By Operator
keys: '2008-04-08' (type: string)
mode: mergepartial
outputColumnNames: _col0
Statistics: Num rows: 500 Data size: 5312 Basic stats: COMPLETE
Column stats: NONE
Select Operator
Statistics: Num rows: 500 Data size: 5312 Basic stats:
COMPLETE Column stats: NONE
Reduce Output Operator
key expressions: '2008-04-08' (type: string)
sort order: +
Map-reduce partition columns: '2008-04-08' (type: string)
Statistics: Num rows: 500 Data size: 5312 Basic stats:
COMPLETE Column stats: NONE
Reducer 3
Reduce Operator Tree:
Join Operator
condition map:
Inner Join 0 to 1
keys:
0 ds (type: string)
1 _col0 (type: string)
Statistics: Num rows: 2200 Data size: 23372 Basic stats:
COMPLETE Column stats: NONE
Group By Operator
aggregations: count()
mode: hash
outputColumnNames: _col0
Statistics: Num rows: 1 Data size: 8 Basic stats: COMPLETE
Column stats: NONE
Reduce Output Operator
sort order:
Statistics: Num rows: 1 Data size: 8 Basic stats: COMPLETE
Column stats: NONE
value expressions: _col0 (type: bigint)
Reducer 4
Reduce Operator Tree:
Group By Operator
aggregations: count(VALUE._col0)
mode: mergepartial
outputColumnNames: _col0
Statistics: Num rows: 1 Data size: 8 Basic stats: COMPLETE
Column stats: NONE
File Output Operator
compressed: false
Statistics: Num rows: 1 Data size: 8 Basic stats: COMPLETE
Column stats: NONE
table:
input format:
org.apache.hadoop.mapred.SequenceFileInputFormat
output format:
org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
Stage: Stage-0
Fetch Operator
limit: -1
Processor Tree:
ListSink
{noformat}
There's an extra MapWork8 in stage-2. The reason is because the joining key is
partition key for both tables. So we created two DPP operators for the query.
Tez doesn't have this problem because it has
{{TezCompiler::runCycleAnalysisForPartitionPruning}} to remove such cyclic DPP
dependency. I think we need to to the same, but not sure why we didn't see the
issue when we implemented the feature. I'll find more about it.
[~csun], [~xuefuz] any ideas?
> Fix and re-enable the spark-only tests
> --------------------------------------
>
> Key: HIVE-15357
> URL: https://issues.apache.org/jira/browse/HIVE-15357
> Project: Hive
> Issue Type: Test
> Reporter: Rui Li
> Assignee: Rui Li
>
> Defined by {{spark.only.query.files}}.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
