You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@impala.apache.org by ta...@apache.org on 2019/01/12 18:19:08 UTC
[09/26] impala git commit: IMPALA-8021: Add estimated cardinality to
EXPLAIN output
http://git-wip-us.apache.org/repos/asf/impala/blob/a7ea86b7/testdata/workloads/functional-planner/queries/PlannerTest/topn.test
----------------------------------------------------------------------
diff --git a/testdata/workloads/functional-planner/queries/PlannerTest/topn.test b/testdata/workloads/functional-planner/queries/PlannerTest/topn.test
index d3ab315..5ff642c 100644
--- a/testdata/workloads/functional-planner/queries/PlannerTest/topn.test
+++ b/testdata/workloads/functional-planner/queries/PlannerTest/topn.test
@@ -7,9 +7,11 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=1]
| order by: name ASC
+| row-size=16B cardinality=0
|
00:SCAN HDFS [functional.testtbl]
partitions=1/1 files=0 size=0B
+ row-size=16B cardinality=0
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -19,9 +21,11 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=1]
| order by: name ASC
+| row-size=16B cardinality=0
|
00:SCAN HDFS [functional.testtbl]
partitions=1/1 files=0 size=0B
+ row-size=16B cardinality=0
====
select zip, count(*)
from functional.testtbl
@@ -34,14 +38,17 @@ PLAN-ROOT SINK
|
02:TOP-N [LIMIT=10]
| order by: count(*) DESC
+| row-size=12B cardinality=0
|
01:AGGREGATE [FINALIZE]
| output: count(*)
| group by: zip
+| row-size=12B cardinality=0
|
00:SCAN HDFS [functional.testtbl]
partitions=1/1 files=0 size=0B
predicates: name LIKE 'm%'
+ row-size=16B cardinality=0
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -51,20 +58,24 @@ PLAN-ROOT SINK
|
02:TOP-N [LIMIT=10]
| order by: count(*) DESC
+| row-size=12B cardinality=0
|
04:AGGREGATE [FINALIZE]
| output: count:merge(*)
| group by: zip
+| row-size=12B cardinality=0
|
03:EXCHANGE [HASH(zip)]
|
01:AGGREGATE [STREAMING]
| output: count(*)
| group by: zip
+| row-size=12B cardinality=0
|
00:SCAN HDFS [functional.testtbl]
partitions=1/1 files=0 size=0B
predicates: name LIKE 'm%'
+ row-size=16B cardinality=0
====
select int_col, sum(float_col)
from functional_hbase.alltypessmall
@@ -77,13 +88,16 @@ PLAN-ROOT SINK
|
02:TOP-N [LIMIT=4]
| order by: sum(float_col) ASC
+| row-size=12B cardinality=4
|
01:AGGREGATE [FINALIZE]
| output: sum(float_col)
| group by: int_col
+| row-size=12B cardinality=5
|
00:SCAN HBASE [functional_hbase.alltypessmall]
predicates: id < 5
+ row-size=12B cardinality=5
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -93,19 +107,23 @@ PLAN-ROOT SINK
|
02:TOP-N [LIMIT=4]
| order by: sum(float_col) ASC
+| row-size=12B cardinality=4
|
04:AGGREGATE [FINALIZE]
| output: sum:merge(float_col)
| group by: int_col
+| row-size=12B cardinality=5
|
03:EXCHANGE [HASH(int_col)]
|
01:AGGREGATE [STREAMING]
| output: sum(float_col)
| group by: int_col
+| row-size=12B cardinality=5
|
00:SCAN HBASE [functional_hbase.alltypessmall]
predicates: id < 5
+ row-size=12B cardinality=5
====
select int_col, sum(float_col), min(float_col)
from functional_hbase.alltypessmall
@@ -134,18 +152,22 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=2]
| order by: int_col ASC
+| row-size=4B cardinality=2
|
02:HASH JOIN [INNER JOIN]
| hash predicates: t1.id = t2.id
| runtime filters: RF000 <- t2.id
+| row-size=16B cardinality=10
|
|--01:SCAN HDFS [functional.alltypessmall t2]
| partitions=4/4 files=4 size=6.32KB
| predicates: t2.int_col IS NOT NULL
+| row-size=8B cardinality=10
|
00:SCAN HDFS [functional.alltypessmall t1]
partitions=4/4 files=4 size=6.32KB
runtime filters: RF000 -> t1.id
+ row-size=8B cardinality=100
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -155,20 +177,24 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=2]
| order by: int_col ASC
+| row-size=4B cardinality=2
|
02:HASH JOIN [INNER JOIN, BROADCAST]
| hash predicates: t1.id = t2.id
| runtime filters: RF000 <- t2.id
+| row-size=16B cardinality=10
|
|--04:EXCHANGE [BROADCAST]
| |
| 01:SCAN HDFS [functional.alltypessmall t2]
| partitions=4/4 files=4 size=6.32KB
| predicates: t2.int_col IS NOT NULL
+| row-size=8B cardinality=10
|
00:SCAN HDFS [functional.alltypessmall t1]
partitions=4/4 files=4 size=6.32KB
runtime filters: RF000 -> t1.id
+ row-size=8B cardinality=100
====
# Test that the top-n is on int_col and not on the id column
select int_col as id from functional.alltypessmall order by id limit 2
@@ -177,9 +203,11 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=2]
| order by: int_col ASC
+| row-size=4B cardinality=2
|
00:SCAN HDFS [functional.alltypessmall]
partitions=4/4 files=4 size=6.32KB
+ row-size=4B cardinality=100
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -189,9 +217,11 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=2]
| order by: int_col ASC
+| row-size=4B cardinality=2
|
00:SCAN HDFS [functional.alltypessmall]
partitions=4/4 files=4 size=6.32KB
+ row-size=4B cardinality=100
====
# Test that the top-n is on id and not on int_col
select int_col as id from functional.alltypessmall order by functional.alltypessmall.id limit 2
@@ -200,9 +230,11 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=2]
| order by: id ASC
+| row-size=8B cardinality=2
|
00:SCAN HDFS [functional.alltypessmall]
partitions=4/4 files=4 size=6.32KB
+ row-size=8B cardinality=100
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -212,9 +244,11 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=2]
| order by: id ASC
+| row-size=8B cardinality=2
|
00:SCAN HDFS [functional.alltypessmall]
partitions=4/4 files=4 size=6.32KB
+ row-size=8B cardinality=100
====
# Test that the limit/offset is propagated correctly to child nodes. The TOP-N node
# should have the limit/offset specified in the query. Child sort nodes should have a
@@ -228,18 +262,22 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=10 OFFSET=5]
| order by: int_col ASC
+| row-size=4B cardinality=10
|
02:HASH JOIN [INNER JOIN]
| hash predicates: t1.id = t2.id
| runtime filters: RF000 <- t2.id
+| row-size=16B cardinality=10
|
|--01:SCAN HDFS [functional.alltypessmall t2]
| partitions=4/4 files=4 size=6.32KB
| predicates: t2.int_col IS NOT NULL
+| row-size=8B cardinality=10
|
00:SCAN HDFS [functional.alltypessmall t1]
partitions=4/4 files=4 size=6.32KB
runtime filters: RF000 -> t1.id
+ row-size=8B cardinality=100
---- SCANRANGELOCATIONS
NODE 0:
HDFS SPLIT hdfs://localhost:20500/test-warehouse/alltypessmall/year=2009/month=1/090101.txt 0:1610
@@ -261,20 +299,24 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=15]
| order by: int_col ASC
+| row-size=4B cardinality=10
|
02:HASH JOIN [INNER JOIN, BROADCAST]
| hash predicates: t1.id = t2.id
| runtime filters: RF000 <- t2.id
+| row-size=16B cardinality=10
|
|--04:EXCHANGE [BROADCAST]
| |
| 01:SCAN HDFS [functional.alltypessmall t2]
| partitions=4/4 files=4 size=6.32KB
| predicates: t2.int_col IS NOT NULL
+| row-size=8B cardinality=10
|
00:SCAN HDFS [functional.alltypessmall t1]
partitions=4/4 files=4 size=6.32KB
runtime filters: RF000 -> t1.id
+ row-size=8B cardinality=100
====
# test distributed top-n over a union (IMPALA-831)
select int_col, bigint_col from
@@ -287,15 +329,19 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=10 OFFSET=5]
| order by: int_col DESC
+| row-size=12B cardinality=10
|
00:UNION
| pass-through-operands: all
+| row-size=12B cardinality=7.40K
|
|--02:SCAN HDFS [functional.alltypessmall]
| partitions=4/4 files=4 size=6.32KB
+| row-size=12B cardinality=100
|
01:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
+ row-size=12B cardinality=7.30K
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -306,15 +352,19 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=15]
| order by: int_col DESC
+| row-size=12B cardinality=15
|
00:UNION
| pass-through-operands: all
+| row-size=12B cardinality=7.40K
|
|--02:SCAN HDFS [functional.alltypessmall]
| partitions=4/4 files=4 size=6.32KB
+| row-size=12B cardinality=100
|
01:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
+ row-size=12B cardinality=7.30K
====
# test distributed top-n over a union distinct (IMPALA-831)
select int_col, bigint_col from
@@ -327,18 +377,23 @@ PLAN-ROOT SINK
|
04:TOP-N [LIMIT=10 OFFSET=5]
| order by: int_col DESC
+| row-size=12B cardinality=10
|
03:AGGREGATE [FINALIZE]
| group by: id, bool_col, tinyint_col, smallint_col, int_col, bigint_col, float_col, double_col, date_string_col, string_col, timestamp_col, year, month
+| row-size=89B cardinality=7.40K
|
00:UNION
| pass-through-operands: all
+| row-size=89B cardinality=7.40K
|
|--02:SCAN HDFS [functional.alltypessmall]
| partitions=4/4 files=4 size=6.32KB
+| row-size=89B cardinality=100
|
01:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
+ row-size=89B cardinality=7.30K
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
@@ -349,23 +404,29 @@ PLAN-ROOT SINK
|
04:TOP-N [LIMIT=15]
| order by: int_col DESC
+| row-size=12B cardinality=15
|
06:AGGREGATE [FINALIZE]
| group by: id, bool_col, tinyint_col, smallint_col, int_col, bigint_col, float_col, double_col, date_string_col, string_col, timestamp_col, year, month
+| row-size=89B cardinality=7.40K
|
05:EXCHANGE [HASH(id,bool_col,tinyint_col,smallint_col,int_col,bigint_col,float_col,double_col,date_string_col,string_col,timestamp_col,year,month)]
|
03:AGGREGATE [STREAMING]
| group by: id, bool_col, tinyint_col, smallint_col, int_col, bigint_col, float_col, double_col, date_string_col, string_col, timestamp_col, year, month
+| row-size=89B cardinality=7.40K
|
00:UNION
| pass-through-operands: all
+| row-size=89B cardinality=7.40K
|
|--02:SCAN HDFS [functional.alltypessmall]
| partitions=4/4 files=4 size=6.32KB
+| row-size=89B cardinality=100
|
01:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
+ row-size=89B cardinality=7.30K
====
# test that top-n is not placed below an unpartitioned exchange with a limit
select * from (select * from functional.alltypes limit 10) t
@@ -375,15 +436,18 @@ PLAN-ROOT SINK
|
01:TOP-N [LIMIT=10]
| order by: int_col ASC
+| row-size=89B cardinality=10
|
00:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
limit: 10
+ row-size=89B cardinality=10
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
01:TOP-N [LIMIT=10]
| order by: int_col ASC
+| row-size=89B cardinality=10
|
02:EXCHANGE [UNPARTITIONED]
| limit: 10
@@ -391,6 +455,7 @@ PLAN-ROOT SINK
00:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
limit: 10
+ row-size=89B cardinality=10
====
# test that top-n is not placed below an unpartitioned exchange with a limit
select * from
@@ -403,21 +468,26 @@ PLAN-ROOT SINK
|
03:TOP-N [LIMIT=10]
| order by: int_col ASC
+| row-size=89B cardinality=10
|
00:UNION
| pass-through-operands: all
| limit: 10
+| row-size=89B cardinality=10
|
|--02:SCAN HDFS [functional.alltypessmall]
| partitions=4/4 files=4 size=6.32KB
+| row-size=89B cardinality=100
|
01:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
+ row-size=89B cardinality=7.30K
---- DISTRIBUTEDPLAN
PLAN-ROOT SINK
|
03:TOP-N [LIMIT=10]
| order by: int_col ASC
+| row-size=89B cardinality=10
|
04:EXCHANGE [UNPARTITIONED]
| limit: 10
@@ -425,10 +495,13 @@ PLAN-ROOT SINK
00:UNION
| pass-through-operands: all
| limit: 10
+| row-size=89B cardinality=10
|
|--02:SCAN HDFS [functional.alltypessmall]
| partitions=4/4 files=4 size=6.32KB
+| row-size=89B cardinality=100
|
01:SCAN HDFS [functional.alltypes]
partitions=24/24 files=24 size=478.45KB
+ row-size=89B cardinality=7.30K
====