hive-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From pxi...@apache.org
Subject [15/24] hive git commit: HIVE-16996: Add HLL as an alternative to FM sketch to compute stats (Pengcheng Xiong, reviewed by Ashutosh Chauhan, Prasanth Jayachandran)
Date Sat, 15 Jul 2017 08:06:54 GMT
http://git-wip-us.apache.org/repos/asf/hive/blob/b883d313/ql/src/test/results/clientpositive/llap/column_table_stats.q.out
----------------------------------------------------------------------
diff --git a/ql/src/test/results/clientpositive/llap/column_table_stats.q.out b/ql/src/test/results/clientpositive/llap/column_table_stats.q.out
index fb04ee8..c7726fe 100644
--- a/ql/src/test/results/clientpositive/llap/column_table_stats.q.out
+++ b/ql/src/test/results/clientpositive/llap/column_table_stats.q.out
@@ -81,16 +81,16 @@ STAGE PLANS:
                     outputColumnNames: key, value
                     Statistics: Num rows: 29 Data size: 5812 Basic stats: COMPLETE Column stats: NONE
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       mode: hash
                       outputColumnNames: _col0, _col1
-                      Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                      Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
                       Reduce Output Operator
                         null sort order: 
                         sort order: 
-                        Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                        Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
                         tag: -1
-                        value expressions: _col0 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col1 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col0 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col1 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: false
             Execution mode: llap
             LLAP IO: no inputs
@@ -151,13 +151,13 @@ STAGE PLANS:
                 aggregations: compute_stats(VALUE._col0), compute_stats(VALUE._col1)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
                 File Output Operator
                   compressed: false
                   GlobalTableId: 0
 #### A masked pattern was here ####
                   NumFilesPerFileSink: 1
-                  Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                  Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
 #### A masked pattern was here ####
                   table:
                       input format: org.apache.hadoop.mapred.SequenceFileInputFormat
@@ -340,19 +340,19 @@ STAGE PLANS:
                     outputColumnNames: key, value, ds, hr
                     Statistics: Num rows: 58 Data size: 32968 Basic stats: COMPLETE Column stats: PARTIAL
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       keys: ds (type: string), hr (type: string)
                       mode: hash
                       outputColumnNames: _col0, _col1, _col2, _col3
-                      Statistics: Num rows: 2 Data size: 2704 Basic stats: COMPLETE Column stats: PARTIAL
+                      Statistics: Num rows: 2 Data size: 2656 Basic stats: COMPLETE Column stats: PARTIAL
                       Reduce Output Operator
                         key expressions: _col0 (type: string), _col1 (type: string)
                         null sort order: aa
                         sort order: ++
                         Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                        Statistics: Num rows: 2 Data size: 2704 Basic stats: COMPLETE Column stats: PARTIAL
+                        Statistics: Num rows: 2 Data size: 2656 Basic stats: COMPLETE Column stats: PARTIAL
                         tag: -1
-                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: true
             Execution mode: llap
             LLAP IO: no inputs
@@ -757,19 +757,19 @@ STAGE PLANS:
                     outputColumnNames: key, value, ds, hr
                     Statistics: Num rows: 58 Data size: 32968 Basic stats: COMPLETE Column stats: PARTIAL
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       keys: ds (type: string), hr (type: string)
                       mode: hash
                       outputColumnNames: _col0, _col1, _col2, _col3
-                      Statistics: Num rows: 2 Data size: 2704 Basic stats: COMPLETE Column stats: PARTIAL
+                      Statistics: Num rows: 2 Data size: 2656 Basic stats: COMPLETE Column stats: PARTIAL
                       Reduce Output Operator
                         key expressions: _col0 (type: string), _col1 (type: string)
                         null sort order: aa
                         sort order: ++
                         Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                        Statistics: Num rows: 2 Data size: 2704 Basic stats: COMPLETE Column stats: PARTIAL
+                        Statistics: Num rows: 2 Data size: 2656 Basic stats: COMPLETE Column stats: PARTIAL
                         tag: -1
-                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: true
             Execution mode: llap
             LLAP IO: no inputs
@@ -1174,19 +1174,19 @@ STAGE PLANS:
                     outputColumnNames: key, value, ds
                     Statistics: Num rows: 29 Data size: 11148 Basic stats: COMPLETE Column stats: PARTIAL
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       keys: ds (type: string), '11' (type: string)
                       mode: hash
                       outputColumnNames: _col0, _col1, _col2, _col3
-                      Statistics: Num rows: 1 Data size: 1254 Basic stats: COMPLETE Column stats: PARTIAL
+                      Statistics: Num rows: 1 Data size: 1230 Basic stats: COMPLETE Column stats: PARTIAL
                       Reduce Output Operator
                         key expressions: _col0 (type: string), '11' (type: string)
                         null sort order: aa
                         sort order: ++
                         Map-reduce partition columns: _col0 (type: string), '11' (type: string)
-                        Statistics: Num rows: 1 Data size: 1254 Basic stats: COMPLETE Column stats: PARTIAL
+                        Statistics: Num rows: 1 Data size: 1230 Basic stats: COMPLETE Column stats: PARTIAL
                         tag: -1
-                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: true
             Execution mode: llap
             LLAP IO: no inputs

http://git-wip-us.apache.org/repos/asf/hive/blob/b883d313/ql/src/test/results/clientpositive/llap/column_table_stats_orc.q.out
----------------------------------------------------------------------
diff --git a/ql/src/test/results/clientpositive/llap/column_table_stats_orc.q.out b/ql/src/test/results/clientpositive/llap/column_table_stats_orc.q.out
index d55cf30..6dff50f 100644
--- a/ql/src/test/results/clientpositive/llap/column_table_stats_orc.q.out
+++ b/ql/src/test/results/clientpositive/llap/column_table_stats_orc.q.out
@@ -82,16 +82,16 @@ STAGE PLANS:
                     outputColumnNames: key, value
                     Statistics: Num rows: 1 Data size: 170 Basic stats: COMPLETE Column stats: NONE
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       mode: hash
                       outputColumnNames: _col0, _col1
-                      Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                      Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
                       Reduce Output Operator
                         null sort order: 
                         sort order: 
-                        Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                        Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
                         tag: -1
-                        value expressions: _col0 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col1 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col0 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col1 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: false
             Execution mode: llap
             LLAP IO: all inputs
@@ -154,13 +154,13 @@ STAGE PLANS:
                 aggregations: compute_stats(VALUE._col0), compute_stats(VALUE._col1)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
                 File Output Operator
                   compressed: false
                   GlobalTableId: 0
 #### A masked pattern was here ####
                   NumFilesPerFileSink: 1
-                  Statistics: Num rows: 1 Data size: 984 Basic stats: COMPLETE Column stats: NONE
+                  Statistics: Num rows: 1 Data size: 960 Basic stats: COMPLETE Column stats: NONE
 #### A masked pattern was here ####
                   table:
                       input format: org.apache.hadoop.mapred.SequenceFileInputFormat
@@ -337,19 +337,19 @@ STAGE PLANS:
                     outputColumnNames: key, value, ds, hr
                     Statistics: Num rows: 2 Data size: 1076 Basic stats: COMPLETE Column stats: PARTIAL
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       keys: ds (type: string), hr (type: string)
                       mode: hash
                       outputColumnNames: _col0, _col1, _col2, _col3
-                      Statistics: Num rows: 1 Data size: 1352 Basic stats: COMPLETE Column stats: PARTIAL
+                      Statistics: Num rows: 1 Data size: 1328 Basic stats: COMPLETE Column stats: PARTIAL
                       Reduce Output Operator
                         key expressions: _col0 (type: string), _col1 (type: string)
                         null sort order: aa
                         sort order: ++
                         Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                        Statistics: Num rows: 1 Data size: 1352 Basic stats: COMPLETE Column stats: PARTIAL
+                        Statistics: Num rows: 1 Data size: 1328 Basic stats: COMPLETE Column stats: PARTIAL
                         tag: -1
-                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: true
             Execution mode: llap
             LLAP IO: all inputs
@@ -742,19 +742,19 @@ STAGE PLANS:
                     outputColumnNames: key, value, ds
                     Statistics: Num rows: 1 Data size: 354 Basic stats: COMPLETE Column stats: PARTIAL
                     Group By Operator
-                      aggregations: compute_stats(key, 16), compute_stats(value, 16)
+                      aggregations: compute_stats(key, 'hll'), compute_stats(value, 'hll')
                       keys: ds (type: string), '11' (type: string)
                       mode: hash
                       outputColumnNames: _col0, _col1, _col2, _col3
-                      Statistics: Num rows: 1 Data size: 1254 Basic stats: COMPLETE Column stats: PARTIAL
+                      Statistics: Num rows: 1 Data size: 1230 Basic stats: COMPLETE Column stats: PARTIAL
                       Reduce Output Operator
                         key expressions: _col0 (type: string), '11' (type: string)
                         null sort order: aa
                         sort order: ++
                         Map-reduce partition columns: _col0 (type: string), '11' (type: string)
-                        Statistics: Num rows: 1 Data size: 1254 Basic stats: COMPLETE Column stats: PARTIAL
+                        Statistics: Num rows: 1 Data size: 1230 Basic stats: COMPLETE Column stats: PARTIAL
                         tag: -1
-                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string,numbitvectors:int>)
+                        value expressions: _col2 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>), _col3 (type: struct<columntype:string,maxlength:bigint,sumlength:bigint,count:bigint,countnulls:bigint,bitvector:string>)
                         auto parallelism: true
             Execution mode: llap
             LLAP IO: all inputs

http://git-wip-us.apache.org/repos/asf/hive/blob/b883d313/ql/src/test/results/clientpositive/llap/columnstats_part_coltype.q.out
----------------------------------------------------------------------
diff --git a/ql/src/test/results/clientpositive/llap/columnstats_part_coltype.q.out b/ql/src/test/results/clientpositive/llap/columnstats_part_coltype.q.out
index dc50fb7..5e64743 100644
--- a/ql/src/test/results/clientpositive/llap/columnstats_part_coltype.q.out
+++ b/ql/src/test/results/clientpositive/llap/columnstats_part_coltype.q.out
@@ -82,7 +82,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-key                 	int                 	27                  	484                 	0                   	18                  	                    	                    	                    	                    	from deserializer   
+key                 	int                 	27                  	484                 	0                   	20                  	                    	                    	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-02', hr=2, part='partA') value
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -91,7 +91,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	18                  	6.8                 	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	20                  	6.8                 	7                   	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-02', hr=2, part='partB') key
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -136,7 +136,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-key                 	int                 	27                  	484                 	0                   	18                  	                    	                    	                    	                    	from deserializer   
+key                 	int                 	27                  	484                 	0                   	20                  	                    	                    	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-02', hr=2, part='partB') value
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -145,7 +145,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	18                  	6.8                 	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	20                  	6.8                 	7                   	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-02', hr=3, part='partA') key
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -194,7 +194,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-key                 	int                 	27                  	495                 	0                   	28                  	                    	                    	                    	                    	from deserializer   
+key                 	int                 	27                  	495                 	0                   	30                  	                    	                    	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-02', hr=3, part='partA') value
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -203,7 +203,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	18                  	6.833333333333333   	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	30                  	6.833333333333333   	7                   	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-03', hr=3, part='partA') key
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -278,7 +278,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-key                 	int                 	15                  	495                 	0                   	43                  	                    	                    	                    	                    	from deserializer   
+key                 	int                 	15                  	495                 	0                   	40                  	                    	                    	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-03', hr=3, part='partA') value
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -287,7 +287,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	34                  	6.825               	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	40                  	6.825               	7                   	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-03', hr=3, part='partB') key
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -296,7 +296,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-key                 	int                 	15                  	495                 	0                   	51                  	                    	                    	                    	                    	from deserializer   
+key                 	int                 	15                  	495                 	0                   	58                  	                    	                    	                    	                    	from deserializer   
 PREHOOK: query: describe formatted partcolstats partition (ds=date '2015-04-03', hr=3, part='partB') value
 PREHOOK: type: DESCTABLE
 PREHOOK: Input: default@partcolstats
@@ -305,7 +305,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstats
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	53                  	6.883333333333334   	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	58                  	6.883333333333334   	7                   	                    	                    	from deserializer   
 PREHOOK: query: drop table partcolstats
 PREHOOK: type: DROPTABLE
 PREHOOK: Input: default@partcolstats
@@ -358,7 +358,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstatsnum
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	18                  	6.833333333333333   	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	30                  	6.833333333333333   	7                   	                    	                    	from deserializer   
 PREHOOK: query: drop table partcolstatsnum
 PREHOOK: type: DROPTABLE
 PREHOOK: Input: default@partcolstatsnum
@@ -411,7 +411,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstatsdec
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	18                  	6.833333333333333   	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	30                  	6.833333333333333   	7                   	                    	                    	from deserializer   
 PREHOOK: query: drop table partcolstatsdec
 PREHOOK: type: DROPTABLE
 PREHOOK: Input: default@partcolstatsdec
@@ -464,7 +464,7 @@ POSTHOOK: type: DESCTABLE
 POSTHOOK: Input: default@partcolstatschar
 # col_name            	data_type           	min                 	max                 	num_nulls           	distinct_count      	avg_col_len         	max_col_len         	num_trues           	num_falses          	comment             
 	 	 	 	 	 	 	 	 	 	 
-value               	string              	                    	                    	0                   	18                  	6.833333333333333   	7                   	                    	                    	from deserializer   
+value               	string              	                    	                    	0                   	30                  	6.833333333333333   	7                   	                    	                    	from deserializer   
 PREHOOK: query: drop table partcolstatschar
 PREHOOK: type: DROPTABLE
 PREHOOK: Input: default@partcolstatschar

http://git-wip-us.apache.org/repos/asf/hive/blob/b883d313/ql/src/test/results/clientpositive/llap/correlationoptimizer1.q.out
----------------------------------------------------------------------
diff --git a/ql/src/test/results/clientpositive/llap/correlationoptimizer1.q.out b/ql/src/test/results/clientpositive/llap/correlationoptimizer1.q.out
index fab5c9c..1e4e084 100644
--- a/ql/src/test/results/clientpositive/llap/correlationoptimizer1.q.out
+++ b/ql/src/test/results/clientpositive/llap/correlationoptimizer1.q.out
@@ -72,18 +72,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 60 Data size: 5160 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 3440 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -93,11 +93,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -220,18 +220,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 60 Data size: 5160 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 3440 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -241,11 +241,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -360,18 +360,18 @@ STAGE PLANS:
                         outputColumnNames: _col0
                         input vertices:
                           0 Map 1
-                        Statistics: Num rows: 60 Data size: 5160 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 40 Data size: 3440 Basic stats: COMPLETE Column stats: COMPLETE
                         Group By Operator
                           aggregations: count()
                           keys: _col0 (type: string)
                           mode: hash
                           outputColumnNames: _col0, _col1
-                          Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                           Reduce Output Operator
                             key expressions: _col0 (type: string)
                             sort order: +
                             Map-reduce partition columns: _col0 (type: string)
-                            Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                            Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                             value expressions: _col1 (type: bigint)
             Execution mode: llap
             LLAP IO: no inputs
@@ -383,11 +383,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -497,12 +497,12 @@ STAGE PLANS:
                         keys: _col0 (type: string)
                         mode: hash
                         outputColumnNames: _col0
-                        Statistics: Num rows: 205 Data size: 17835 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 250 Data size: 21750 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           key expressions: _col0 (type: string)
                           sort order: +
                           Map-reduce partition columns: _col0 (type: string)
-                          Statistics: Num rows: 205 Data size: 17835 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 250 Data size: 21750 Basic stats: COMPLETE Column stats: COMPLETE
             Execution mode: llap
             LLAP IO: no inputs
         Reducer 2 
@@ -515,18 +515,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 25 Data size: 2150 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 20 Data size: 1720 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -536,11 +536,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -650,12 +650,12 @@ STAGE PLANS:
                         keys: _col0 (type: string)
                         mode: hash
                         outputColumnNames: _col0
-                        Statistics: Num rows: 205 Data size: 17835 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 250 Data size: 21750 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           key expressions: _col0 (type: string)
                           sort order: +
                           Map-reduce partition columns: _col0 (type: string)
-                          Statistics: Num rows: 205 Data size: 17835 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 250 Data size: 21750 Basic stats: COMPLETE Column stats: COMPLETE
             Execution mode: llap
             LLAP IO: no inputs
         Reducer 2 
@@ -668,18 +668,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 25 Data size: 2150 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 20 Data size: 1720 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -689,11 +689,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 12 Data size: 1128 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 10 Data size: 940 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -810,18 +810,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 60 Data size: 5160 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 3440 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -831,11 +831,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -952,18 +952,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 60 Data size: 5160 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 3440 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -973,11 +973,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1316 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1504 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -1094,7 +1094,7 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col1
-                Statistics: Num rows: 60 Data size: 5220 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 3480 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col1 (type: string)
@@ -1236,7 +1236,7 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col1
-                Statistics: Num rows: 60 Data size: 5220 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 3480 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col1 (type: string)
@@ -1646,12 +1646,12 @@ STAGE PLANS:
                   keys: _col1 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -1661,11 +1661,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -1788,12 +1788,12 @@ STAGE PLANS:
                   keys: _col1 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -1803,11 +1803,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 14 Data size: 1330 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 16 Data size: 1520 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -1924,18 +1924,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 500 Data size: 5246 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 500 Data size: 3526 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -1945,11 +1945,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -2066,18 +2066,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 500 Data size: 5246 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 500 Data size: 3526 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -2087,11 +2087,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -2208,18 +2208,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 525 Data size: 5246 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 525 Data size: 3526 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -2229,11 +2229,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -2350,18 +2350,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0
-                Statistics: Num rows: 525 Data size: 5246 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 525 Data size: 3526 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -2371,11 +2371,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 15 Data size: 378 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 17 Data size: 308 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1)
                     mode: hash
@@ -2499,18 +2499,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 60 Data size: 10500 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 7000 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string), _col1 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1, _col2
-                  Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string), _col1 (type: string)
                     sort order: ++
                     Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                    Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col2 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -2520,11 +2520,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2
-                Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int), hash(_col2) (type: int)
                   outputColumnNames: _col0, _col1, _col2
-                  Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1), sum(_col2)
                     mode: hash
@@ -2648,18 +2648,18 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0, _col1
-                Statistics: Num rows: 60 Data size: 10500 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 40 Data size: 7000 Basic stats: COMPLETE Column stats: COMPLETE
                 Group By Operator
                   aggregations: count()
                   keys: _col0 (type: string), _col1 (type: string)
                   mode: hash
                   outputColumnNames: _col0, _col1, _col2
-                  Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string), _col1 (type: string)
                     sort order: ++
                     Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                    Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col2 (type: bigint)
         Reducer 3 
             Execution mode: llap
@@ -2669,11 +2669,11 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2
-                Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: hash(_col0) (type: int), hash(_col1) (type: int), hash(_col2) (type: int)
                   outputColumnNames: _col0, _col1, _col2
-                  Statistics: Num rows: 30 Data size: 5490 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 20 Data size: 3660 Basic stats: COMPLETE Column stats: COMPLETE
                   Group By Operator
                     aggregations: sum(_col0), sum(_col1), sum(_col2)
                     mode: hash


Mime
View raw message