You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hive.apache.org by dj...@apache.org on 2018/02/15 20:27:04 UTC
hive git commit: HIVE-18721 : Bucket Map Join : Handle empty buckets
(Deepak Jaiswal, reviewed by Gunther Hagleitner)
Repository: hive
Updated Branches:
refs/heads/master dabb62d6e -> 217811254
HIVE-18721 : Bucket Map Join : Handle empty buckets (Deepak Jaiswal, reviewed by Gunther Hagleitner)
Project: http://git-wip-us.apache.org/repos/asf/hive/repo
Commit: http://git-wip-us.apache.org/repos/asf/hive/commit/21781125
Tree: http://git-wip-us.apache.org/repos/asf/hive/tree/21781125
Diff: http://git-wip-us.apache.org/repos/asf/hive/diff/21781125
Branch: refs/heads/master
Commit: 21781125419826bd1ae76d980f958419a84d84f3
Parents: dabb62d
Author: Deepak Jaiswal <dj...@apache.org>
Authored: Thu Feb 15 12:26:07 2018 -0800
Committer: Deepak Jaiswal <dj...@apache.org>
Committed: Thu Feb 15 12:26:54 2018 -0800
----------------------------------------------------------------------
.../test/resources/testconfiguration.properties | 1 +
.../hive/ql/exec/tez/CustomPartitionEdge.java | 7 ++
.../clientpositive/bucket_map_join_tez_empty.q | 18 +++
.../llap/bucket_map_join_tez_empty.q.out | 121 +++++++++++++++++++
4 files changed, 147 insertions(+)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/hive/blob/21781125/itests/src/test/resources/testconfiguration.properties
----------------------------------------------------------------------
diff --git a/itests/src/test/resources/testconfiguration.properties b/itests/src/test/resources/testconfiguration.properties
index c2252f3..942f97a 100644
--- a/itests/src/test/resources/testconfiguration.properties
+++ b/itests/src/test/resources/testconfiguration.properties
@@ -506,6 +506,7 @@ minillaplocal.query.files=\
bucket_many.q,\
bucket_map_join_tez1.q,\
bucket_map_join_tez2.q,\
+ bucket_map_join_tez_empty.q,\
bucketizedhiveinputformat.q,\
bucketmapjoin6.q,\
bucketmapjoin7.q,\
http://git-wip-us.apache.org/repos/asf/hive/blob/21781125/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/CustomPartitionEdge.java
----------------------------------------------------------------------
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/CustomPartitionEdge.java b/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/CustomPartitionEdge.java
index 1ac1d14..4248cd9 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/CustomPartitionEdge.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/CustomPartitionEdge.java
@@ -20,6 +20,7 @@ package org.apache.hadoop.hive.ql.exec.tez;
import java.io.IOException;
import java.nio.ByteBuffer;
+import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
import java.util.Map;
@@ -84,6 +85,12 @@ public class CustomPartitionEdge extends EdgeManagerPlugin {
@Override
public void routeDataMovementEventToDestination(DataMovementEvent event,
int sourceTaskIndex, int sourceOutputIndex, Map<Integer, List<Integer>> mapDestTaskIndices) {
+ if (conf.getRoutingTable().get(sourceOutputIndex).size() == 0) {
+ // No task for given input, return empty list with -1 as index
+ mapDestTaskIndices.put(-1, new ArrayList<>());
+ return;
+ }
+ // Normal case.
List<Integer> outputIndices = Collections.singletonList(sourceTaskIndex);
for (Integer destIndex : conf.getRoutingTable().get(sourceOutputIndex)) {
mapDestTaskIndices.put(destIndex, outputIndices);
http://git-wip-us.apache.org/repos/asf/hive/blob/21781125/ql/src/test/queries/clientpositive/bucket_map_join_tez_empty.q
----------------------------------------------------------------------
diff --git a/ql/src/test/queries/clientpositive/bucket_map_join_tez_empty.q b/ql/src/test/queries/clientpositive/bucket_map_join_tez_empty.q
new file mode 100644
index 0000000..cc43b5b
--- /dev/null
+++ b/ql/src/test/queries/clientpositive/bucket_map_join_tez_empty.q
@@ -0,0 +1,18 @@
+set hive.mapred.mode=nonstrict;
+set hive.explain.user=false;
+set hive.auto.convert.join=true;
+set hive.auto.convert.join.noconditionaltask=true;
+set hive.auto.convert.join.noconditionaltask.size=10000;
+
+CREATE TABLE tab1(key1 int, value string) CLUSTERED BY (key1) INTO 10 BUCKETS STORED AS TEXTFILE;
+CREATE TABLE tab2 (key1 int, value string) CLUSTERED BY (key1) INTO 10 BUCKETS STORED AS TEXTFILE;
+
+
+-- HIVE-18721 : Make sure only certain buckets have data.
+insert into tab1 VALUES (1,"abc"),(4,"def"),(8, "ghi");
+insert into tab2 VALUES (1, "abc"), (5, "aa");
+
+set hive.convert.join.bucket.mapjoin.tez = true;
+
+explain select * from tab1, tab2 where tab1.key1 = tab2.key1;
+select * from tab1, tab2 where tab1.key1 = tab2.key1;
http://git-wip-us.apache.org/repos/asf/hive/blob/21781125/ql/src/test/results/clientpositive/llap/bucket_map_join_tez_empty.q.out
----------------------------------------------------------------------
diff --git a/ql/src/test/results/clientpositive/llap/bucket_map_join_tez_empty.q.out b/ql/src/test/results/clientpositive/llap/bucket_map_join_tez_empty.q.out
new file mode 100644
index 0000000..33825da
--- /dev/null
+++ b/ql/src/test/results/clientpositive/llap/bucket_map_join_tez_empty.q.out
@@ -0,0 +1,121 @@
+PREHOOK: query: CREATE TABLE tab1(key1 int, value string) CLUSTERED BY (key1) INTO 10 BUCKETS STORED AS TEXTFILE
+PREHOOK: type: CREATETABLE
+PREHOOK: Output: database:default
+PREHOOK: Output: default@tab1
+POSTHOOK: query: CREATE TABLE tab1(key1 int, value string) CLUSTERED BY (key1) INTO 10 BUCKETS STORED AS TEXTFILE
+POSTHOOK: type: CREATETABLE
+POSTHOOK: Output: database:default
+POSTHOOK: Output: default@tab1
+PREHOOK: query: CREATE TABLE tab2 (key1 int, value string) CLUSTERED BY (key1) INTO 10 BUCKETS STORED AS TEXTFILE
+PREHOOK: type: CREATETABLE
+PREHOOK: Output: database:default
+PREHOOK: Output: default@tab2
+POSTHOOK: query: CREATE TABLE tab2 (key1 int, value string) CLUSTERED BY (key1) INTO 10 BUCKETS STORED AS TEXTFILE
+POSTHOOK: type: CREATETABLE
+POSTHOOK: Output: database:default
+POSTHOOK: Output: default@tab2
+PREHOOK: query: insert into tab1 VALUES (1,"abc"),(4,"def"),(8, "ghi")
+PREHOOK: type: QUERY
+PREHOOK: Input: _dummy_database@_dummy_table
+PREHOOK: Output: default@tab1
+POSTHOOK: query: insert into tab1 VALUES (1,"abc"),(4,"def"),(8, "ghi")
+POSTHOOK: type: QUERY
+POSTHOOK: Input: _dummy_database@_dummy_table
+POSTHOOK: Output: default@tab1
+POSTHOOK: Lineage: tab1.key1 SCRIPT []
+POSTHOOK: Lineage: tab1.value SCRIPT []
+PREHOOK: query: insert into tab2 VALUES (1, "abc"), (5, "aa")
+PREHOOK: type: QUERY
+PREHOOK: Input: _dummy_database@_dummy_table
+PREHOOK: Output: default@tab2
+POSTHOOK: query: insert into tab2 VALUES (1, "abc"), (5, "aa")
+POSTHOOK: type: QUERY
+POSTHOOK: Input: _dummy_database@_dummy_table
+POSTHOOK: Output: default@tab2
+POSTHOOK: Lineage: tab2.key1 SCRIPT []
+POSTHOOK: Lineage: tab2.value SCRIPT []
+PREHOOK: query: explain select * from tab1, tab2 where tab1.key1 = tab2.key1
+PREHOOK: type: QUERY
+POSTHOOK: query: explain select * from tab1, tab2 where tab1.key1 = tab2.key1
+POSTHOOK: type: QUERY
+STAGE DEPENDENCIES:
+ Stage-1 is a root stage
+ Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+ Stage: Stage-1
+ Tez
+#### A masked pattern was here ####
+ Edges:
+ Map 1 <- Map 2 (CUSTOM_EDGE)
+#### A masked pattern was here ####
+ Vertices:
+ Map 1
+ Map Operator Tree:
+ TableScan
+ alias: tab1
+ Statistics: Num rows: 3 Data size: 273 Basic stats: COMPLETE Column stats: COMPLETE
+ Filter Operator
+ predicate: key1 is not null (type: boolean)
+ Statistics: Num rows: 3 Data size: 273 Basic stats: COMPLETE Column stats: COMPLETE
+ Select Operator
+ expressions: key1 (type: int), value (type: string)
+ outputColumnNames: _col0, _col1
+ Statistics: Num rows: 3 Data size: 273 Basic stats: COMPLETE Column stats: COMPLETE
+ Map Join Operator
+ condition map:
+ Inner Join 0 to 1
+ keys:
+ 0 _col0 (type: int)
+ 1 _col0 (type: int)
+ outputColumnNames: _col0, _col1, _col2, _col3
+ input vertices:
+ 1 Map 2
+ Statistics: Num rows: 2 Data size: 364 Basic stats: COMPLETE Column stats: COMPLETE
+ File Output Operator
+ compressed: false
+ Statistics: Num rows: 2 Data size: 364 Basic stats: COMPLETE Column stats: COMPLETE
+ 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
+ Execution mode: llap
+ LLAP IO: no inputs
+ Map 2
+ Map Operator Tree:
+ TableScan
+ alias: tab2
+ Statistics: Num rows: 2 Data size: 182 Basic stats: COMPLETE Column stats: COMPLETE
+ Filter Operator
+ predicate: key1 is not null (type: boolean)
+ Statistics: Num rows: 2 Data size: 182 Basic stats: COMPLETE Column stats: COMPLETE
+ Select Operator
+ expressions: key1 (type: int), value (type: string)
+ outputColumnNames: _col0, _col1
+ Statistics: Num rows: 2 Data size: 182 Basic stats: COMPLETE Column stats: COMPLETE
+ Reduce Output Operator
+ key expressions: _col0 (type: int)
+ sort order: +
+ Map-reduce partition columns: _col0 (type: int)
+ Statistics: Num rows: 2 Data size: 182 Basic stats: COMPLETE Column stats: COMPLETE
+ value expressions: _col1 (type: string)
+ Execution mode: llap
+ LLAP IO: no inputs
+
+ Stage: Stage-0
+ Fetch Operator
+ limit: -1
+ Processor Tree:
+ ListSink
+
+PREHOOK: query: select * from tab1, tab2 where tab1.key1 = tab2.key1
+PREHOOK: type: QUERY
+PREHOOK: Input: default@tab1
+PREHOOK: Input: default@tab2
+#### A masked pattern was here ####
+POSTHOOK: query: select * from tab1, tab2 where tab1.key1 = tab2.key1
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@tab1
+POSTHOOK: Input: default@tab2
+#### A masked pattern was here ####
+1 abc 1 abc