http://git-wip-us.apache.org/repos/asf/hive/blob/86b18772/ql/src/test/results/clientpositive/perf/query54.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/query54.q.out b/ql/src/test/results/clientpositive/perf/query54.q.out index e1f5135..3cbcbe3 100644 --- a/ql/src/test/results/clientpositive/perf/query54.q.out +++ b/ql/src/test/results/clientpositive/perf/query54.q.out @@ -1,175 +1,396 @@ -PREHOOK: query: explain with my_customers as ( select c_customer_sk , c_current_addr_sk from ( select cs_sold_date_sk sold_date_sk, cs_bill_customer_sk customer_sk, cs_item_sk item_sk from catalog_sales union all select ws_sold_date_sk sold_date_sk, ws_bill_customer_sk customer_sk, ws_item_sk item_sk from web_sales ) cs_or_ws_sales, item, date_dim, customer where sold_date_sk = d_date_sk and item_sk = i_item_sk and i_category = 'Jewelry' and i_class = 'football' and c_customer_sk = cs_or_ws_sales.customer_sk and d_moy = 3 and d_year = 2000 group by c_customer_sk , c_current_addr_sk ) , my_revenue as ( select c_customer_sk, sum(ss_ext_sales_price) as revenue from my_customers, store_sales, customer_address, store, date_dim where c_current_addr_sk = ca_address_sk and ca_county = s_county and ca_state = s_state and ss_sold_date_sk = d_date_sk and c_customer_sk = ss_customer_sk and d_month_seq between (1203) and (1205) group by c_customer_sk ) , segments as (select cast((revenue/50) as int) as segment from my_revenue ) select segment, count(*) as num_customers, segment*50 as segment_base from segments group by segment order by segment, num_customers limit 100 +Warning: Shuffle Join MERGEJOIN[191][tables = [$hdt$_1, $hdt$_2]] in Stage 'Reducer 3' is a cross product +Warning: Shuffle Join MERGEJOIN[188][tables = [$hdt$_1, $hdt$_2, $hdt$_3, $hdt$_0, $hdt$_4]] in Stage 'Reducer 12' is a cross product +Warning: Shuffle Join MERGEJOIN[190][tables = [$hdt$_1, $hdt$_2, $hdt$_3, $hdt$_0, $hdt$_4, $hdt$_5, $hdt$_6]] in Stage 'Reducer 14' is a cross product +Warning: Shuffle Join MERGEJOIN[192][tables = [$hdt$_1, $hdt$_2, $hdt$_0]] in Stage 'Reducer 4' is a cross product +PREHOOK: query: explain +with my_customers as ( + select distinct c_customer_sk + , c_current_addr_sk + from + ( select cs_sold_date_sk sold_date_sk, + cs_bill_customer_sk customer_sk, + cs_item_sk item_sk + from catalog_sales + union all + select ws_sold_date_sk sold_date_sk, + ws_bill_customer_sk customer_sk, + ws_item_sk item_sk + from web_sales + ) cs_or_ws_sales, + item, + date_dim, + customer + where sold_date_sk = d_date_sk + and item_sk = i_item_sk + and i_category = 'Jewelry' + and i_class = 'consignment' + and c_customer_sk = cs_or_ws_sales.customer_sk + and d_moy = 3 + and d_year = 1999 + ) + , my_revenue as ( + select c_customer_sk, + sum(ss_ext_sales_price) as revenue + from my_customers, + store_sales, + customer_address, + store, + date_dim + where c_current_addr_sk = ca_address_sk + and ca_county = s_county + and ca_state = s_state + and ss_sold_date_sk = d_date_sk + and c_customer_sk = ss_customer_sk + and d_month_seq between (select distinct d_month_seq+1 + from date_dim where d_year = 1999 and d_moy = 3) + and (select distinct d_month_seq+3 + from date_dim where d_year = 1999 and d_moy = 3) + group by c_customer_sk + ) + , segments as + (select cast((revenue/50) as int) as segment + from my_revenue + ) + select segment, count(*) as num_customers, segment*50 as segment_base + from segments + group by segment + order by segment, num_customers + limit 100 PREHOOK: type: QUERY -POSTHOOK: query: explain with my_customers as ( select c_customer_sk , c_current_addr_sk from ( select cs_sold_date_sk sold_date_sk, cs_bill_customer_sk customer_sk, cs_item_sk item_sk from catalog_sales union all select ws_sold_date_sk sold_date_sk, ws_bill_customer_sk customer_sk, ws_item_sk item_sk from web_sales ) cs_or_ws_sales, item, date_dim, customer where sold_date_sk = d_date_sk and item_sk = i_item_sk and i_category = 'Jewelry' and i_class = 'football' and c_customer_sk = cs_or_ws_sales.customer_sk and d_moy = 3 and d_year = 2000 group by c_customer_sk , c_current_addr_sk ) , my_revenue as ( select c_customer_sk, sum(ss_ext_sales_price) as revenue from my_customers, store_sales, customer_address, store, date_dim where c_current_addr_sk = ca_address_sk and ca_county = s_county and ca_state = s_state and ss_sold_date_sk = d_date_sk and c_customer_sk = ss_customer_sk and d_month_seq between (1203) and (1205) group by c_customer_sk ) , segments as (select cast((revenue/50) as int) as segment from my_revenue ) select segment, count(*) as num_customers, segment*50 as segment_base from segments group by segment order by segment, num_customers limit 100 +POSTHOOK: query: explain +with my_customers as ( + select distinct c_customer_sk + , c_current_addr_sk + from + ( select cs_sold_date_sk sold_date_sk, + cs_bill_customer_sk customer_sk, + cs_item_sk item_sk + from catalog_sales + union all + select ws_sold_date_sk sold_date_sk, + ws_bill_customer_sk customer_sk, + ws_item_sk item_sk + from web_sales + ) cs_or_ws_sales, + item, + date_dim, + customer + where sold_date_sk = d_date_sk + and item_sk = i_item_sk + and i_category = 'Jewelry' + and i_class = 'consignment' + and c_customer_sk = cs_or_ws_sales.customer_sk + and d_moy = 3 + and d_year = 1999 + ) + , my_revenue as ( + select c_customer_sk, + sum(ss_ext_sales_price) as revenue + from my_customers, + store_sales, + customer_address, + store, + date_dim + where c_current_addr_sk = ca_address_sk + and ca_county = s_county + and ca_state = s_state + and ss_sold_date_sk = d_date_sk + and c_customer_sk = ss_customer_sk + and d_month_seq between (select distinct d_month_seq+1 + from date_dim where d_year = 1999 and d_moy = 3) + and (select distinct d_month_seq+3 + from date_dim where d_year = 1999 and d_moy = 3) + group by c_customer_sk + ) + , segments as + (select cast((revenue/50) as int) as segment + from my_revenue + ) + select segment, count(*) as num_customers, segment*50 as segment_base + from segments + group by segment + order by segment, num_customers + limit 100 POSTHOOK: type: QUERY Plan optimized by CBO. Vertex dependency in root stage -Map 12 <- Union 13 (CONTAINS) -Map 18 <- Union 13 (CONTAINS) -Reducer 10 <- Reducer 17 (SIMPLE_EDGE), Reducer 9 (SIMPLE_EDGE) -Reducer 14 <- Map 19 (SIMPLE_EDGE), Union 13 (SIMPLE_EDGE) -Reducer 15 <- Map 20 (SIMPLE_EDGE), Reducer 14 (SIMPLE_EDGE) -Reducer 16 <- Map 21 (SIMPLE_EDGE), Reducer 15 (SIMPLE_EDGE) -Reducer 17 <- Reducer 16 (SIMPLE_EDGE) -Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 7 (SIMPLE_EDGE) -Reducer 3 <- Reducer 10 (SIMPLE_EDGE), Reducer 2 (SIMPLE_EDGE) -Reducer 4 <- Reducer 3 (SIMPLE_EDGE) +Map 21 <- Union 22 (CONTAINS) +Map 27 <- Union 22 (CONTAINS) +Reducer 10 <- Map 1 (SIMPLE_EDGE) +Reducer 11 <- Reducer 10 (CUSTOM_SIMPLE_EDGE) +Reducer 12 <- Reducer 11 (CUSTOM_SIMPLE_EDGE), Reducer 19 (CUSTOM_SIMPLE_EDGE) +Reducer 13 <- Map 32 (SIMPLE_EDGE), Reducer 12 (SIMPLE_EDGE) +Reducer 14 <- Reducer 13 (CUSTOM_SIMPLE_EDGE), Reducer 15 (CUSTOM_SIMPLE_EDGE) +Reducer 15 <- Map 1 (SIMPLE_EDGE) +Reducer 17 <- Map 16 (SIMPLE_EDGE), Map 20 (SIMPLE_EDGE) +Reducer 18 <- Reducer 17 (SIMPLE_EDGE), Reducer 26 (SIMPLE_EDGE) +Reducer 19 <- Map 31 (SIMPLE_EDGE), Reducer 18 (SIMPLE_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE) +Reducer 23 <- Map 28 (SIMPLE_EDGE), Union 22 (SIMPLE_EDGE) +Reducer 24 <- Map 29 (SIMPLE_EDGE), Reducer 23 (SIMPLE_EDGE) +Reducer 25 <- Map 30 (SIMPLE_EDGE), Reducer 24 (SIMPLE_EDGE) +Reducer 26 <- Reducer 25 (SIMPLE_EDGE) +Reducer 3 <- Reducer 2 (CUSTOM_SIMPLE_EDGE), Reducer 9 (CUSTOM_SIMPLE_EDGE) +Reducer 4 <- Reducer 14 (CUSTOM_SIMPLE_EDGE), Reducer 3 (CUSTOM_SIMPLE_EDGE) Reducer 5 <- Reducer 4 (SIMPLE_EDGE) Reducer 6 <- Reducer 5 (SIMPLE_EDGE) -Reducer 9 <- Map 11 (SIMPLE_EDGE), Map 8 (SIMPLE_EDGE) +Reducer 7 <- Reducer 6 (SIMPLE_EDGE) +Reducer 8 <- Map 1 (SIMPLE_EDGE) +Reducer 9 <- Reducer 8 (CUSTOM_SIMPLE_EDGE) Stage-0 Fetch Operator limit:100 Stage-1 - Reducer 6 - File Output Operator [FS_69] - Limit [LIM_68] (rows=100 width=88) + Reducer 7 + File Output Operator [FS_129] + Limit [LIM_128] (rows=100 width=158) Number of rows:100 - Select Operator [SEL_67] (rows=174238687 width=88) + Select Operator [SEL_127] (rows=1614130953450400 width=158) Output:["_col0","_col1","_col2"] - <-Reducer 5 [SIMPLE_EDGE] - SHUFFLE [RS_66] - Select Operator [SEL_65] (rows=174238687 width=88) + <-Reducer 6 [SIMPLE_EDGE] + SHUFFLE [RS_126] + Select Operator [SEL_125] (rows=1614130953450400 width=158) Output:["_col0","_col1","_col2"] - Group By Operator [GBY_64] (rows=174238687 width=88) + Group By Operator [GBY_124] (rows=1614130953450400 width=158) Output:["_col0","_col1"],aggregations:["count(VALUE._col0)"],keys:KEY._col0 - <-Reducer 4 [SIMPLE_EDGE] - SHUFFLE [RS_63] + <-Reducer 5 [SIMPLE_EDGE] + SHUFFLE [RS_123] PartitionCols:_col0 - Group By Operator [GBY_62] (rows=348477374 width=88) + Group By Operator [GBY_122] (rows=3228261906900801 width=158) Output:["_col0","_col1"],aggregations:["count()"],keys:_col0 - Select Operator [SEL_60] (rows=348477374 width=88) + Select Operator [SEL_120] (rows=3228261906900801 width=158) Output:["_col0"] - Group By Operator [GBY_59] (rows=348477374 width=88) + Group By Operator [GBY_119] (rows=3228261906900801 width=158) Output:["_col0","_col1"],aggregations:["sum(VALUE._col0)"],keys:KEY._col0 - <-Reducer 3 [SIMPLE_EDGE] - SHUFFLE [RS_58] + <-Reducer 4 [SIMPLE_EDGE] + SHUFFLE [RS_118] PartitionCols:_col0 - Group By Operator [GBY_57] (rows=696954748 width=88) - Output:["_col0","_col1"],aggregations:["sum(_col2)"],keys:_col10 - Merge Join Operator [MERGEJOIN_117] (rows=696954748 width=88) - Conds:RS_53._col1=RS_54._col5(Inner),Output:["_col2","_col10"] - <-Reducer 10 [SIMPLE_EDGE] - SHUFFLE [RS_54] - PartitionCols:_col5 - Merge Join Operator [MERGEJOIN_116] (rows=316240138 width=135) - Conds:RS_46._col0=RS_47._col1(Inner),Output:["_col5"] - <-Reducer 17 [SIMPLE_EDGE] - SHUFFLE [RS_47] - PartitionCols:_col1 - Select Operator [SEL_42] (rows=287491029 width=135) - Output:["_col0","_col1"] - Group By Operator [GBY_41] (rows=287491029 width=135) - Output:["_col0","_col1"],keys:KEY._col0, KEY._col1 - <-Reducer 16 [SIMPLE_EDGE] - SHUFFLE [RS_40] - PartitionCols:_col0, _col1 - Group By Operator [GBY_39] (rows=574982058 width=135) - Output:["_col0","_col1"],keys:_col10, _col9 - Merge Join Operator [MERGEJOIN_115] (rows=574982058 width=135) - Conds:RS_35._col1=RS_36._col0(Inner),Output:["_col9","_col10"] - <-Map 21 [SIMPLE_EDGE] - SHUFFLE [RS_36] + Group By Operator [GBY_117] (rows=6456523813801603 width=158) + Output:["_col0","_col1"],aggregations:["sum(_col4)"],keys:_col0 + Select Operator [SEL_116] (rows=6456523813801603 width=158) + Output:["_col0","_col4"] + Filter Operator [FIL_115] (rows=6456523813801603 width=158) + predicate:_col11 BETWEEN _col13 AND _col15 + Select Operator [SEL_114] (rows=58108714324214428 width=158) + Output:["_col0","_col4","_col11","_col13","_col15"] + Merge Join Operator [MERGEJOIN_192] (rows=58108714324214428 width=158) + Conds:(Inner),Output:["_col0","_col2","_col6","_col13","_col15"] + <-Reducer 14 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_112] + Select Operator [SEL_107] (rows=6363893803988 width=1217) + Output:["_col0","_col4","_col11","_col13"] + Merge Join Operator [MERGEJOIN_190] (rows=6363893803988 width=1217) + Conds:(Left Outer),Output:["_col5","_col9","_col12","_col13"] + <-Reducer 13 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_104] + Merge Join Operator [MERGEJOIN_189] (rows=696954748 width=97) + Conds:RS_101._col7=RS_102._col0(Inner),Output:["_col5","_col9","_col12"] + <-Map 32 [SIMPLE_EDGE] + SHUFFLE [RS_102] PartitionCols:_col0 - Select Operator [SEL_28] (rows=80000000 width=860) + Select Operator [SEL_80] (rows=73049 width=1119) Output:["_col0","_col1"] - Filter Operator [FIL_110] (rows=80000000 width=860) - predicate:(c_customer_sk is not null and c_current_addr_sk is not null) - TableScan [TS_26] (rows=80000000 width=860) - default@customer,customer,Tbl:COMPLETE,Col:NONE,Output:["c_customer_sk","c_current_addr_sk"] - <-Reducer 15 [SIMPLE_EDGE] - SHUFFLE [RS_35] - PartitionCols:_col1 - Merge Join Operator [MERGEJOIN_114] (rows=522710951 width=135) - Conds:RS_32._col2=RS_33._col0(Inner),Output:["_col1"] - <-Map 20 [SIMPLE_EDGE] - SHUFFLE [RS_33] - PartitionCols:_col0 - Select Operator [SEL_25] (rows=115500 width=1436) - Output:["_col0"] - Filter Operator [FIL_109] (rows=115500 width=1436) - predicate:((i_category = 'Jewelry') and (i_class = 'football') and i_item_sk is not null) - TableScan [TS_23] (rows=462000 width=1436) - default@item,item,Tbl:COMPLETE,Col:NONE,Output:["i_item_sk","i_class","i_category"] - <-Reducer 14 [SIMPLE_EDGE] - SHUFFLE [RS_32] - PartitionCols:_col2 - Merge Join Operator [MERGEJOIN_113] (rows=475191764 width=135) - Conds:Union 13._col0=RS_30._col0(Inner),Output:["_col1","_col2"] - <-Map 19 [SIMPLE_EDGE] - SHUFFLE [RS_30] - PartitionCols:_col0 - Select Operator [SEL_22] (rows=18262 width=1119) - Output:["_col0"] - Filter Operator [FIL_108] (rows=18262 width=1119) - predicate:((d_moy = 3) and (d_year = 2000) and d_date_sk is not null) - TableScan [TS_20] (rows=73049 width=1119) - default@date_dim,date_dim,Tbl:COMPLETE,Col:NONE,Output:["d_date_sk","d_year","d_moy"] - <-Union 13 [SIMPLE_EDGE] - <-Map 12 [CONTAINS] - Reduce Output Operator [RS_29] - PartitionCols:_col0 - Select Operator [SEL_14] (rows=287989836 width=135) - Output:["_col0","_col1","_col2"] - Filter Operator [FIL_106] (rows=287989836 width=135) - predicate:(cs_item_sk is not null and cs_sold_date_sk is not null and cs_bill_customer_sk is not null) - TableScan [TS_12] (rows=287989836 width=135) - Output:["cs_sold_date_sk","cs_bill_customer_sk","cs_item_sk"] - <-Map 18 [CONTAINS] - Reduce Output Operator [RS_29] + Filter Operator [FIL_180] (rows=73049 width=1119) + predicate:d_date_sk is not null + TableScan [TS_78] (rows=73049 width=1119) + default@date_dim,date_dim,Tbl:COMPLETE,Col:NONE,Output:["d_date_sk","d_month_seq"] + <-Reducer 12 [SIMPLE_EDGE] + SHUFFLE [RS_101] + PartitionCols:_col7 + Merge Join Operator [MERGEJOIN_188] (rows=633595212 width=97) + Conds:(Inner),Output:["_col5","_col7","_col9"] + <-Reducer 11 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_99] + Select Operator [SEL_77] (rows=1 width=8) + Filter Operator [FIL_76] (rows=1 width=8) + predicate:(sq_count_check(_col0) <= 1) + Group By Operator [GBY_74] (rows=1 width=8) + Output:["_col0"],aggregations:["count(VALUE._col0)"] + <-Reducer 10 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_73] + Group By Operator [GBY_72] (rows=1 width=8) + Output:["_col0"],aggregations:["count()"] + Select Operator [SEL_70] (rows=9131 width=1119) + Group By Operator [GBY_69] (rows=9131 width=1119) + Output:["_col0"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] + SHUFFLE [RS_68] + PartitionCols:_col0 + Group By Operator [GBY_67] (rows=18262 width=1119) + Output:["_col0"],keys:_col0 + Select Operator [SEL_65] (rows=18262 width=1119) + Output:["_col0"] + Filter Operator [FIL_179] (rows=18262 width=1119) + predicate:((d_year = 1999) and (d_moy = 3)) + TableScan [TS_0] (rows=73049 width=1119) + default@date_dim,date_dim,Tbl:COMPLETE,Col:NONE,Output:["d_month_seq","d_year","d_moy"] + <-Reducer 19 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_98] + Merge Join Operator [MERGEJOIN_187] (rows=633595212 width=88) + Conds:RS_95._col5=RS_96._col1(Inner),Output:["_col5","_col7","_col9"] + <-Map 31 [SIMPLE_EDGE] + SHUFFLE [RS_96] + PartitionCols:_col1 + Select Operator [SEL_62] (rows=575995635 width=88) + Output:["_col0","_col1","_col2"] + Filter Operator [FIL_178] (rows=575995635 width=88) + predicate:(ss_customer_sk is not null and ss_sold_date_sk is not null) + TableScan [TS_60] (rows=575995635 width=88) + default@store_sales,store_sales,Tbl:COMPLETE,Col:NONE,Output:["ss_sold_date_sk","ss_customer_sk","ss_ext_sales_price"] + <-Reducer 18 [SIMPLE_EDGE] + SHUFFLE [RS_95] + PartitionCols:_col5 + Merge Join Operator [MERGEJOIN_186] (rows=316240138 width=135) + Conds:RS_92._col0=RS_93._col1(Inner),Output:["_col5"] + <-Reducer 17 [SIMPLE_EDGE] + SHUFFLE [RS_92] + PartitionCols:_col0 + Merge Join Operator [MERGEJOIN_182] (rows=44000000 width=1014) + Conds:RS_89._col1, _col2=RS_90._col0, _col1(Inner),Output:["_col0"] + <-Map 16 [SIMPLE_EDGE] + SHUFFLE [RS_89] + PartitionCols:_col1, _col2 + Select Operator [SEL_25] (rows=40000000 width=1014) + Output:["_col0","_col1","_col2"] + Filter Operator [FIL_171] (rows=40000000 width=1014) + predicate:(ca_address_sk is not null and ca_county is not null and ca_state is not null) + TableScan [TS_23] (rows=40000000 width=1014) + default@customer_address,customer_address,Tbl:COMPLETE,Col:NONE,Output:["ca_address_sk","ca_county","ca_state"] + <-Map 20 [SIMPLE_EDGE] + SHUFFLE [RS_90] + PartitionCols:_col0, _col1 + Select Operator [SEL_28] (rows=1704 width=1910) + Output:["_col0","_col1"] + Filter Operator [FIL_172] (rows=1704 width=1910) + predicate:(s_county is not null and s_state is not null) + TableScan [TS_26] (rows=1704 width=1910) + default@store,store,Tbl:COMPLETE,Col:NONE,Output:["s_county","s_state"] + <-Reducer 26 [SIMPLE_EDGE] + SHUFFLE [RS_93] + PartitionCols:_col1 + Select Operator [SEL_59] (rows=287491029 width=135) + Output:["_col0","_col1"] + Group By Operator [GBY_58] (rows=287491029 width=135) + Output:["_col0","_col1"],keys:KEY._col0, KEY._col1 + <-Reducer 25 [SIMPLE_EDGE] + SHUFFLE [RS_57] + PartitionCols:_col0, _col1 + Group By Operator [GBY_56] (rows=574982058 width=135) + Output:["_col0","_col1"],keys:_col10, _col9 + Merge Join Operator [MERGEJOIN_185] (rows=574982058 width=135) + Conds:RS_52._col1=RS_53._col0(Inner),Output:["_col9","_col10"] + <-Map 30 [SIMPLE_EDGE] + SHUFFLE [RS_53] + PartitionCols:_col0 + Select Operator [SEL_45] (rows=80000000 width=860) + Output:["_col0","_col1"] + Filter Operator [FIL_177] (rows=80000000 width=860) + predicate:(c_customer_sk is not null and c_current_addr_sk is not null) + TableScan [TS_43] (rows=80000000 width=860) + default@customer,customer,Tbl:COMPLETE,Col:NONE,Output:["c_customer_sk","c_current_addr_sk"] + <-Reducer 24 [SIMPLE_EDGE] + SHUFFLE [RS_52] + PartitionCols:_col1 + Merge Join Operator [MERGEJOIN_184] (rows=522710951 width=135) + Conds:RS_49._col2=RS_50._col0(Inner),Output:["_col1"] + <-Map 29 [SIMPLE_EDGE] + SHUFFLE [RS_50] + PartitionCols:_col0 + Select Operator [SEL_42] (rows=115500 width=1436) + Output:["_col0"] + Filter Operator [FIL_176] (rows=115500 width=1436) + predicate:((i_category = 'Jewelry') and (i_class = 'consignment') and i_item_sk is not null) + TableScan [TS_40] (rows=462000 width=1436) + default@item,item,Tbl:COMPLETE,Col:NONE,Output:["i_item_sk","i_class","i_category"] + <-Reducer 23 [SIMPLE_EDGE] + SHUFFLE [RS_49] + PartitionCols:_col2 + Merge Join Operator [MERGEJOIN_183] (rows=475191764 width=135) + Conds:Union 22._col0=RS_47._col0(Inner),Output:["_col1","_col2"] + <-Map 28 [SIMPLE_EDGE] + SHUFFLE [RS_47] + PartitionCols:_col0 + Select Operator [SEL_39] (rows=18262 width=1119) + Output:["_col0"] + Filter Operator [FIL_175] (rows=18262 width=1119) + predicate:((d_moy = 3) and (d_year = 1999) and d_date_sk is not null) + TableScan [TS_37] (rows=73049 width=1119) + default@date_dim,date_dim,Tbl:COMPLETE,Col:NONE,Output:["d_date_sk","d_year","d_moy"] + <-Union 22 [SIMPLE_EDGE] + <-Map 21 [CONTAINS] + Reduce Output Operator [RS_46] + PartitionCols:_col0 + Select Operator [SEL_31] (rows=287989836 width=135) + Output:["_col0","_col1","_col2"] + Filter Operator [FIL_173] (rows=287989836 width=135) + predicate:(cs_item_sk is not null and cs_sold_date_sk is not null and cs_bill_customer_sk is not null) + TableScan [TS_29] (rows=287989836 width=135) + Output:["cs_sold_date_sk","cs_bill_customer_sk","cs_item_sk"] + <-Map 27 [CONTAINS] + Reduce Output Operator [RS_46] + PartitionCols:_col0 + Select Operator [SEL_34] (rows=144002668 width=135) + Output:["_col0","_col1","_col2"] + Filter Operator [FIL_174] (rows=144002668 width=135) + predicate:(ws_item_sk is not null and ws_sold_date_sk is not null and ws_bill_customer_sk is not null) + TableScan [TS_32] (rows=144002668 width=135) + Output:["ws_sold_date_sk","ws_item_sk","ws_bill_customer_sk"] + <-Reducer 15 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_105] + Group By Operator [GBY_87] (rows=9131 width=1119) + Output:["_col0"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] + SHUFFLE [RS_86] + PartitionCols:_col0 + Group By Operator [GBY_85] (rows=18262 width=1119) + Output:["_col0"],keys:_col0 + Select Operator [SEL_83] (rows=18262 width=1119) + Output:["_col0"] + Filter Operator [FIL_181] (rows=18262 width=1119) + predicate:((d_year = 1999) and (d_moy = 3)) + Please refer to the previous TableScan [TS_0] + <-Reducer 3 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_111] + Merge Join Operator [MERGEJOIN_191] (rows=9131 width=1128) + Conds:(Right Outer),Output:["_col0"] + <-Reducer 2 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_108] + Group By Operator [GBY_6] (rows=9131 width=1119) + Output:["_col0"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] + SHUFFLE [RS_5] + PartitionCols:_col0 + Group By Operator [GBY_4] (rows=18262 width=1119) + Output:["_col0"],keys:_col0 + Select Operator [SEL_2] (rows=18262 width=1119) + Output:["_col0"] + Filter Operator [FIL_169] (rows=18262 width=1119) + predicate:((d_year = 1999) and (d_moy = 3)) + Please refer to the previous TableScan [TS_0] + <-Reducer 9 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_109] + Select Operator [SEL_22] (rows=1 width=8) + Filter Operator [FIL_21] (rows=1 width=8) + predicate:(sq_count_check(_col0) <= 1) + Group By Operator [GBY_19] (rows=1 width=8) + Output:["_col0"],aggregations:["count(VALUE._col0)"] + <-Reducer 8 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_18] + Group By Operator [GBY_17] (rows=1 width=8) + Output:["_col0"],aggregations:["count()"] + Select Operator [SEL_15] (rows=9131 width=1119) + Group By Operator [GBY_14] (rows=9131 width=1119) + Output:["_col0"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] + SHUFFLE [RS_13] PartitionCols:_col0 - Select Operator [SEL_17] (rows=144002668 width=135) - Output:["_col0","_col1","_col2"] - Filter Operator [FIL_107] (rows=144002668 width=135) - predicate:(ws_item_sk is not null and ws_sold_date_sk is not null and ws_bill_customer_sk is not null) - TableScan [TS_15] (rows=144002668 width=135) - Output:["ws_sold_date_sk","ws_item_sk","ws_bill_customer_sk"] - <-Reducer 9 [SIMPLE_EDGE] - SHUFFLE [RS_46] - PartitionCols:_col0 - Merge Join Operator [MERGEJOIN_112] (rows=44000000 width=1014) - Conds:RS_43._col1, _col2=RS_44._col0, _col1(Inner),Output:["_col0"] - <-Map 11 [SIMPLE_EDGE] - SHUFFLE [RS_44] - PartitionCols:_col0, _col1 - Select Operator [SEL_11] (rows=1704 width=1910) - Output:["_col0","_col1"] - Filter Operator [FIL_105] (rows=1704 width=1910) - predicate:(s_county is not null and s_state is not null) - TableScan [TS_9] (rows=1704 width=1910) - default@store,store,Tbl:COMPLETE,Col:NONE,Output:["s_county","s_state"] - <-Map 8 [SIMPLE_EDGE] - SHUFFLE [RS_43] - PartitionCols:_col1, _col2 - Select Operator [SEL_8] (rows=40000000 width=1014) - Output:["_col0","_col1","_col2"] - Filter Operator [FIL_104] (rows=40000000 width=1014) - predicate:(ca_address_sk is not null and ca_county is not null and ca_state is not null) - TableScan [TS_6] (rows=40000000 width=1014) - default@customer_address,customer_address,Tbl:COMPLETE,Col:NONE,Output:["ca_address_sk","ca_county","ca_state"] - <-Reducer 2 [SIMPLE_EDGE] - SHUFFLE [RS_53] - PartitionCols:_col1 - Merge Join Operator [MERGEJOIN_111] (rows=633595212 width=88) - Conds:RS_50._col0=RS_51._col0(Inner),Output:["_col1","_col2"] - <-Map 1 [SIMPLE_EDGE] - SHUFFLE [RS_50] - PartitionCols:_col0 - Select Operator [SEL_2] (rows=575995635 width=88) - Output:["_col0","_col1","_col2"] - Filter Operator [FIL_102] (rows=575995635 width=88) - predicate:(ss_customer_sk is not null and ss_sold_date_sk is not null) - TableScan [TS_0] (rows=575995635 width=88) - default@store_sales,store_sales,Tbl:COMPLETE,Col:NONE,Output:["ss_sold_date_sk","ss_customer_sk","ss_ext_sales_price"] - <-Map 7 [SIMPLE_EDGE] - SHUFFLE [RS_51] - PartitionCols:_col0 - Select Operator [SEL_5] (rows=8116 width=1119) - Output:["_col0"] - Filter Operator [FIL_103] (rows=8116 width=1119) - predicate:(d_month_seq BETWEEN 1203 AND 1205 and d_date_sk is not null) - TableScan [TS_3] (rows=73049 width=1119) - default@date_dim,date_dim,Tbl:COMPLETE,Col:NONE,Output:["d_date_sk","d_month_seq"] + Group By Operator [GBY_12] (rows=18262 width=1119) + Output:["_col0"],keys:_col0 + Select Operator [SEL_10] (rows=18262 width=1119) + Output:["_col0"] + Filter Operator [FIL_170] (rows=18262 width=1119) + predicate:((d_year = 1999) and (d_moy = 3)) + Please refer to the previous TableScan [TS_0]
http://git-wip-us.apache.org/repos/asf/hive/blob/86b18772/ql/src/test/results/clientpositive/perf/query55.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/query55.q.out b/ql/src/test/results/clientpositive/perf/query55.q.out index 7948b6d..7125422 100644 --- a/ql/src/test/results/clientpositive/perf/query55.q.out +++ b/ql/src/test/results/clientpositive/perf/query55.q.out @@ -1,6 +1,28 @@ -PREHOOK: query: explain select i_brand_id brand_id, i_brand brand, sum(ss_ext_sales_price) ext_price from date_dim, store_sales, item where date_dim.d_date_sk = store_sales.ss_sold_date_sk and store_sales.ss_item_sk = item.i_item_sk and i_manager_id=36 and d_moy=12 and d_year=2001 group by i_brand, i_brand_id order by ext_price desc, i_brand_id limit 100 +PREHOOK: query: explain +select i_brand_id brand_id, i_brand brand, + sum(ss_ext_sales_price) ext_price + from date_dim, store_sales, item + where d_date_sk = ss_sold_date_sk + and ss_item_sk = i_item_sk + and i_manager_id=36 + and d_moy=12 + and d_year=2001 + group by i_brand, i_brand_id + order by ext_price desc, i_brand_id +limit 100 PREHOOK: type: QUERY -POSTHOOK: query: explain select i_brand_id brand_id, i_brand brand, sum(ss_ext_sales_price) ext_price from date_dim, store_sales, item where date_dim.d_date_sk = store_sales.ss_sold_date_sk and store_sales.ss_item_sk = item.i_item_sk and i_manager_id=36 and d_moy=12 and d_year=2001 group by i_brand, i_brand_id order by ext_price desc, i_brand_id limit 100 +POSTHOOK: query: explain +select i_brand_id brand_id, i_brand brand, + sum(ss_ext_sales_price) ext_price + from date_dim, store_sales, item + where d_date_sk = ss_sold_date_sk + and ss_item_sk = i_item_sk + and i_manager_id=36 + and d_moy=12 + and d_year=2001 + group by i_brand, i_brand_id + order by ext_price desc, i_brand_id +limit 100 POSTHOOK: type: QUERY Plan optimized by CBO. http://git-wip-us.apache.org/repos/asf/hive/blob/86b18772/ql/src/test/results/clientpositive/perf/query56.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/query56.q.out b/ql/src/test/results/clientpositive/perf/query56.q.out index 9e710e0..4ec7201 100644 --- a/ql/src/test/results/clientpositive/perf/query56.q.out +++ b/ql/src/test/results/clientpositive/perf/query56.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: explain with ss as ( +PREHOOK: query: explain +with ss as ( select i_item_id,sum(ss_ext_sales_price) total_sales from store_sales, @@ -64,7 +65,8 @@ where i_color in ('orchid','chiffon','lace')) order by total_sales limit 100 PREHOOK: type: QUERY -POSTHOOK: query: explain with ss as ( +POSTHOOK: query: explain +with ss as ( select i_item_id,sum(ss_ext_sales_price) total_sales from store_sales,
