http://git-wip-us.apache.org/repos/asf/hive/blob/b8299551/ql/src/test/results/clientpositive/perf/tez/constraints/query28.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/tez/constraints/query28.q.out b/ql/src/test/results/clientpositive/perf/tez/constraints/query28.q.out new file mode 100644 index 0000000..c7fd970 --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/tez/constraints/query28.q.out @@ -0,0 +1,296 @@ +Warning: Shuffle Join MERGEJOIN[102][tables = [$hdt$_0, $hdt$_1]] in Stage 'Reducer 4' is a cross product +Warning: Shuffle Join MERGEJOIN[103][tables = [$hdt$_0, $hdt$_1, $hdt$_2]] in Stage 'Reducer 5' is a cross product +Warning: Shuffle Join MERGEJOIN[104][tables = [$hdt$_0, $hdt$_1, $hdt$_2, $hdt$_3]] in Stage 'Reducer 6' is a cross product +Warning: Shuffle Join MERGEJOIN[105][tables = [$hdt$_0, $hdt$_1, $hdt$_2, $hdt$_3, $hdt$_4]] in Stage 'Reducer 7' is a cross product +Warning: Shuffle Join MERGEJOIN[106][tables = [$hdt$_0, $hdt$_1, $hdt$_2, $hdt$_3, $hdt$_4, $hdt$_5]] in Stage 'Reducer 8' is a cross product +PREHOOK: query: explain +select * +from (select avg(ss_list_price) B1_LP + ,count(ss_list_price) B1_CNT + ,count(distinct ss_list_price) B1_CNTD + from store_sales + where ss_quantity between 0 and 5 + and (ss_list_price between 11 and 11+10 + or ss_coupon_amt between 460 and 460+1000 + or ss_wholesale_cost between 14 and 14+20)) B1, + (select avg(ss_list_price) B2_LP + ,count(ss_list_price) B2_CNT + ,count(distinct ss_list_price) B2_CNTD + from store_sales + where ss_quantity between 6 and 10 + and (ss_list_price between 91 and 91+10 + or ss_coupon_amt between 1430 and 1430+1000 + or ss_wholesale_cost between 32 and 32+20)) B2, + (select avg(ss_list_price) B3_LP + ,count(ss_list_price) B3_CNT + ,count(distinct ss_list_price) B3_CNTD + from store_sales + where ss_quantity between 11 and 15 + and (ss_list_price between 66 and 66+10 + or ss_coupon_amt between 920 and 920+1000 + or ss_wholesale_cost between 4 and 4+20)) B3, + (select avg(ss_list_price) B4_LP + ,count(ss_list_price) B4_CNT + ,count(distinct ss_list_price) B4_CNTD + from store_sales + where ss_quantity between 16 and 20 + and (ss_list_price between 142 and 142+10 + or ss_coupon_amt between 3054 and 3054+1000 + or ss_wholesale_cost between 80 and 80+20)) B4, + (select avg(ss_list_price) B5_LP + ,count(ss_list_price) B5_CNT + ,count(distinct ss_list_price) B5_CNTD + from store_sales + where ss_quantity between 21 and 25 + and (ss_list_price between 135 and 135+10 + or ss_coupon_amt between 14180 and 14180+1000 + or ss_wholesale_cost between 38 and 38+20)) B5, + (select avg(ss_list_price) B6_LP + ,count(ss_list_price) B6_CNT + ,count(distinct ss_list_price) B6_CNTD + from store_sales + where ss_quantity between 26 and 30 + and (ss_list_price between 28 and 28+10 + or ss_coupon_amt between 2513 and 2513+1000 + or ss_wholesale_cost between 42 and 42+20)) B6 +limit 100 +PREHOOK: type: QUERY +PREHOOK: Input: default@store_sales +PREHOOK: Output: hdfs://### HDFS PATH ### +POSTHOOK: query: explain +select * +from (select avg(ss_list_price) B1_LP + ,count(ss_list_price) B1_CNT + ,count(distinct ss_list_price) B1_CNTD + from store_sales + where ss_quantity between 0 and 5 + and (ss_list_price between 11 and 11+10 + or ss_coupon_amt between 460 and 460+1000 + or ss_wholesale_cost between 14 and 14+20)) B1, + (select avg(ss_list_price) B2_LP + ,count(ss_list_price) B2_CNT + ,count(distinct ss_list_price) B2_CNTD + from store_sales + where ss_quantity between 6 and 10 + and (ss_list_price between 91 and 91+10 + or ss_coupon_amt between 1430 and 1430+1000 + or ss_wholesale_cost between 32 and 32+20)) B2, + (select avg(ss_list_price) B3_LP + ,count(ss_list_price) B3_CNT + ,count(distinct ss_list_price) B3_CNTD + from store_sales + where ss_quantity between 11 and 15 + and (ss_list_price between 66 and 66+10 + or ss_coupon_amt between 920 and 920+1000 + or ss_wholesale_cost between 4 and 4+20)) B3, + (select avg(ss_list_price) B4_LP + ,count(ss_list_price) B4_CNT + ,count(distinct ss_list_price) B4_CNTD + from store_sales + where ss_quantity between 16 and 20 + and (ss_list_price between 142 and 142+10 + or ss_coupon_amt between 3054 and 3054+1000 + or ss_wholesale_cost between 80 and 80+20)) B4, + (select avg(ss_list_price) B5_LP + ,count(ss_list_price) B5_CNT + ,count(distinct ss_list_price) B5_CNTD + from store_sales + where ss_quantity between 21 and 25 + and (ss_list_price between 135 and 135+10 + or ss_coupon_amt between 14180 and 14180+1000 + or ss_wholesale_cost between 38 and 38+20)) B5, + (select avg(ss_list_price) B6_LP + ,count(ss_list_price) B6_CNT + ,count(distinct ss_list_price) B6_CNTD + from store_sales + where ss_quantity between 26 and 30 + and (ss_list_price between 28 and 28+10 + or ss_coupon_amt between 2513 and 2513+1000 + or ss_wholesale_cost between 42 and 42+20)) B6 +limit 100 +POSTHOOK: type: QUERY +POSTHOOK: Input: default@store_sales +POSTHOOK: Output: hdfs://### HDFS PATH ### +Plan optimized by CBO. + +Vertex dependency in root stage +Reducer 10 <- Reducer 9 (CUSTOM_SIMPLE_EDGE) +Reducer 11 <- Map 1 (SIMPLE_EDGE) +Reducer 12 <- Reducer 11 (CUSTOM_SIMPLE_EDGE) +Reducer 13 <- Map 1 (SIMPLE_EDGE) +Reducer 14 <- Reducer 13 (CUSTOM_SIMPLE_EDGE) +Reducer 15 <- Map 1 (SIMPLE_EDGE) +Reducer 16 <- Reducer 15 (CUSTOM_SIMPLE_EDGE) +Reducer 17 <- Map 1 (SIMPLE_EDGE) +Reducer 18 <- Reducer 17 (CUSTOM_SIMPLE_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE) +Reducer 3 <- Reducer 2 (CUSTOM_SIMPLE_EDGE) +Reducer 4 <- Reducer 10 (CUSTOM_SIMPLE_EDGE), Reducer 3 (CUSTOM_SIMPLE_EDGE) +Reducer 5 <- Reducer 12 (CUSTOM_SIMPLE_EDGE), Reducer 4 (CUSTOM_SIMPLE_EDGE) +Reducer 6 <- Reducer 14 (CUSTOM_SIMPLE_EDGE), Reducer 5 (CUSTOM_SIMPLE_EDGE) +Reducer 7 <- Reducer 16 (CUSTOM_SIMPLE_EDGE), Reducer 6 (CUSTOM_SIMPLE_EDGE) +Reducer 8 <- Reducer 18 (CUSTOM_SIMPLE_EDGE), Reducer 7 (CUSTOM_SIMPLE_EDGE) +Reducer 9 <- Map 1 (SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:100 + Stage-1 + Reducer 8 + File Output Operator [FS_59] + Limit [LIM_58] (rows=1 width=768) + Number of rows:100 + Select Operator [SEL_57] (rows=1 width=768) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12","_col13","_col14","_col15","_col16","_col17"] + Merge Join Operator [MERGEJOIN_106] (rows=1 width=768) + Conds:(Inner),Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12","_col13","_col14","_col15","_col16","_col17"] + <-Reducer 18 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_166] + Select Operator [SEL_165] (rows=1 width=128) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_164] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)","count(VALUE._col2)"] + <-Reducer 17 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_163] + Group By Operator [GBY_162] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col1)","count(_col2)","count(_col0)"] + Group By Operator [GBY_161] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_130] + PartitionCols:_col0 + Group By Operator [GBY_124] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(ss_list_price)","count(ss_list_price)"],keys:ss_list_price + Select Operator [SEL_118] (rows=15237978 width=324) + Output:["ss_list_price"] + Filter Operator [FIL_112] (rows=15237978 width=324) + predicate:((ss_list_price BETWEEN 91 AND 101 or ss_coupon_amt BETWEEN 1430 AND 2430 or ss_wholesale_cost BETWEEN 32 AND 52) and ss_quantity BETWEEN 6 AND 10) + TableScan [TS_0] (rows=575995635 width=324) + default@store_sales,store_sales,Tbl:COMPLETE,Col:COMPLETE,Output:["ss_quantity","ss_wholesale_cost","ss_list_price","ss_coupon_amt"] + <-Reducer 7 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_54] + Merge Join Operator [MERGEJOIN_105] (rows=1 width=640) + Conds:(Inner),Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12","_col13","_col14"] + <-Reducer 16 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_160] + Select Operator [SEL_159] (rows=1 width=128) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_158] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)","count(VALUE._col2)"] + <-Reducer 15 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_157] + Group By Operator [GBY_156] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col1)","count(_col2)","count(_col0)"] + Group By Operator [GBY_155] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_129] + PartitionCols:_col0 + Group By Operator [GBY_123] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(ss_list_price)","count(ss_list_price)"],keys:ss_list_price + Select Operator [SEL_117] (rows=15237978 width=324) + Output:["ss_list_price"] + Filter Operator [FIL_111] (rows=15237978 width=324) + predicate:((ss_list_price BETWEEN 66 AND 76 or ss_coupon_amt BETWEEN 920 AND 1920 or ss_wholesale_cost BETWEEN 4 AND 24) and ss_quantity BETWEEN 11 AND 15) + Please refer to the previous TableScan [TS_0] + <-Reducer 6 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_51] + Merge Join Operator [MERGEJOIN_104] (rows=1 width=512) + Conds:(Inner),Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11"] + <-Reducer 14 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_154] + Select Operator [SEL_153] (rows=1 width=128) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_152] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)","count(VALUE._col2)"] + <-Reducer 13 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_151] + Group By Operator [GBY_150] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col1)","count(_col2)","count(_col0)"] + Group By Operator [GBY_149] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_128] + PartitionCols:_col0 + Group By Operator [GBY_122] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(ss_list_price)","count(ss_list_price)"],keys:ss_list_price + Select Operator [SEL_116] (rows=15237978 width=324) + Output:["ss_list_price"] + Filter Operator [FIL_110] (rows=15237978 width=324) + predicate:((ss_list_price BETWEEN 142 AND 152 or ss_coupon_amt BETWEEN 3054 AND 4054 or ss_wholesale_cost BETWEEN 80 AND 100) and ss_quantity BETWEEN 16 AND 20) + Please refer to the previous TableScan [TS_0] + <-Reducer 5 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_48] + Merge Join Operator [MERGEJOIN_103] (rows=1 width=384) + Conds:(Inner),Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8"] + <-Reducer 12 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_148] + Select Operator [SEL_147] (rows=1 width=128) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_146] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)","count(VALUE._col2)"] + <-Reducer 11 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_145] + Group By Operator [GBY_144] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col1)","count(_col2)","count(_col0)"] + Group By Operator [GBY_143] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_127] + PartitionCols:_col0 + Group By Operator [GBY_121] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(ss_list_price)","count(ss_list_price)"],keys:ss_list_price + Select Operator [SEL_115] (rows=15237978 width=324) + Output:["ss_list_price"] + Filter Operator [FIL_109] (rows=15237978 width=324) + predicate:((ss_list_price BETWEEN 135 AND 145 or ss_coupon_amt BETWEEN 14180 AND 15180 or ss_wholesale_cost BETWEEN 38 AND 58) and ss_quantity BETWEEN 21 AND 25) + Please refer to the previous TableScan [TS_0] + <-Reducer 4 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_45] + Merge Join Operator [MERGEJOIN_102] (rows=1 width=256) + Conds:(Inner),Output:["_col0","_col1","_col2","_col3","_col4","_col5"] + <-Reducer 10 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_142] + Select Operator [SEL_141] (rows=1 width=128) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_140] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)","count(VALUE._col2)"] + <-Reducer 9 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_139] + Group By Operator [GBY_138] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col1)","count(_col2)","count(_col0)"] + Group By Operator [GBY_137] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_126] + PartitionCols:_col0 + Group By Operator [GBY_120] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(ss_list_price)","count(ss_list_price)"],keys:ss_list_price + Select Operator [SEL_114] (rows=15237978 width=324) + Output:["ss_list_price"] + Filter Operator [FIL_108] (rows=15237978 width=324) + predicate:((ss_list_price BETWEEN 28 AND 38 or ss_coupon_amt BETWEEN 2513 AND 3513 or ss_wholesale_cost BETWEEN 42 AND 62) and ss_quantity BETWEEN 26 AND 30) + Please refer to the previous TableScan [TS_0] + <-Reducer 3 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_136] + Select Operator [SEL_135] (rows=1 width=128) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_134] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)","count(VALUE._col2)"] + <-Reducer 2 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_133] + Group By Operator [GBY_132] (rows=1 width=128) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col1)","count(_col2)","count(_col0)"] + Group By Operator [GBY_131] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)","count(VALUE._col1)"],keys:KEY._col0 + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_125] + PartitionCols:_col0 + Group By Operator [GBY_119] (rows=7618989 width=226) + Output:["_col0","_col1","_col2"],aggregations:["sum(ss_list_price)","count(ss_list_price)"],keys:ss_list_price + Select Operator [SEL_113] (rows=15237978 width=324) + Output:["ss_list_price"] + Filter Operator [FIL_107] (rows=15237978 width=324) + predicate:((ss_list_price BETWEEN 11 AND 21 or ss_coupon_amt BETWEEN 460 AND 1460 or ss_wholesale_cost BETWEEN 14 AND 34) and ss_quantity BETWEEN 0 AND 5) + Please refer to the previous TableScan [TS_0] +
http://git-wip-us.apache.org/repos/asf/hive/blob/b8299551/ql/src/test/results/clientpositive/perf/tez/constraints/query29.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/tez/constraints/query29.q.out b/ql/src/test/results/clientpositive/perf/tez/constraints/query29.q.out new file mode 100644 index 0000000..715f1ce --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/tez/constraints/query29.q.out @@ -0,0 +1,328 @@ +PREHOOK: query: explain +select + i_item_id + ,i_item_desc + ,s_store_id + ,s_store_name + ,sum(ss_quantity) as store_sales_quantity + ,sum(sr_return_quantity) as store_returns_quantity + ,sum(cs_quantity) as catalog_sales_quantity + from + store_sales + ,store_returns + ,catalog_sales + ,date_dim d1 + ,date_dim d2 + ,date_dim d3 + ,store + ,item + where + d1.d_moy = 4 + and d1.d_year = 1999 + and d1.d_date_sk = ss_sold_date_sk + and i_item_sk = ss_item_sk + and s_store_sk = ss_store_sk + and ss_customer_sk = sr_customer_sk + and ss_item_sk = sr_item_sk + and ss_ticket_number = sr_ticket_number + and sr_returned_date_sk = d2.d_date_sk + and d2.d_moy between 4 and 4 + 3 + and d2.d_year = 1999 + and sr_customer_sk = cs_bill_customer_sk + and sr_item_sk = cs_item_sk + and cs_sold_date_sk = d3.d_date_sk + and d3.d_year in (1999,1999+1,1999+2) + group by + i_item_id + ,i_item_desc + ,s_store_id + ,s_store_name + order by + i_item_id + ,i_item_desc + ,s_store_id + ,s_store_name + limit 100 +PREHOOK: type: QUERY +PREHOOK: Input: default@catalog_sales +PREHOOK: Input: default@date_dim +PREHOOK: Input: default@item +PREHOOK: Input: default@store +PREHOOK: Input: default@store_returns +PREHOOK: Input: default@store_sales +PREHOOK: Output: hdfs://### HDFS PATH ### +POSTHOOK: query: explain +select + i_item_id + ,i_item_desc + ,s_store_id + ,s_store_name + ,sum(ss_quantity) as store_sales_quantity + ,sum(sr_return_quantity) as store_returns_quantity + ,sum(cs_quantity) as catalog_sales_quantity + from + store_sales + ,store_returns + ,catalog_sales + ,date_dim d1 + ,date_dim d2 + ,date_dim d3 + ,store + ,item + where + d1.d_moy = 4 + and d1.d_year = 1999 + and d1.d_date_sk = ss_sold_date_sk + and i_item_sk = ss_item_sk + and s_store_sk = ss_store_sk + and ss_customer_sk = sr_customer_sk + and ss_item_sk = sr_item_sk + and ss_ticket_number = sr_ticket_number + and sr_returned_date_sk = d2.d_date_sk + and d2.d_moy between 4 and 4 + 3 + and d2.d_year = 1999 + and sr_customer_sk = cs_bill_customer_sk + and sr_item_sk = cs_item_sk + and cs_sold_date_sk = d3.d_date_sk + and d3.d_year in (1999,1999+1,1999+2) + group by + i_item_id + ,i_item_desc + ,s_store_id + ,s_store_name + order by + i_item_id + ,i_item_desc + ,s_store_id + ,s_store_name + limit 100 +POSTHOOK: type: QUERY +POSTHOOK: Input: default@catalog_sales +POSTHOOK: Input: default@date_dim +POSTHOOK: Input: default@item +POSTHOOK: Input: default@store +POSTHOOK: Input: default@store_returns +POSTHOOK: Input: default@store_sales +POSTHOOK: Output: hdfs://### HDFS PATH ### +Plan optimized by CBO. + +Vertex dependency in root stage +Map 1 <- Reducer 9 (BROADCAST_EDGE) +Map 10 <- Reducer 16 (BROADCAST_EDGE), Reducer 18 (BROADCAST_EDGE), Reducer 19 (BROADCAST_EDGE), Reducer 20 (BROADCAST_EDGE), Reducer 6 (BROADCAST_EDGE), Reducer 7 (BROADCAST_EDGE) +Reducer 11 <- Map 10 (SIMPLE_EDGE), Map 15 (SIMPLE_EDGE) +Reducer 12 <- Reducer 11 (SIMPLE_EDGE), Reducer 17 (SIMPLE_EDGE) +Reducer 13 <- Map 22 (SIMPLE_EDGE), Reducer 12 (SIMPLE_EDGE) +Reducer 14 <- Map 23 (SIMPLE_EDGE), Reducer 13 (SIMPLE_EDGE) +Reducer 16 <- Map 15 (CUSTOM_SIMPLE_EDGE) +Reducer 17 <- Map 15 (SIMPLE_EDGE), Map 21 (SIMPLE_EDGE) +Reducer 18 <- Reducer 17 (CUSTOM_SIMPLE_EDGE) +Reducer 19 <- Reducer 17 (CUSTOM_SIMPLE_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 8 (SIMPLE_EDGE) +Reducer 20 <- Reducer 17 (CUSTOM_SIMPLE_EDGE) +Reducer 3 <- Reducer 14 (SIMPLE_EDGE), Reducer 2 (SIMPLE_EDGE) +Reducer 4 <- Reducer 3 (SIMPLE_EDGE) +Reducer 5 <- Reducer 4 (SIMPLE_EDGE) +Reducer 6 <- Reducer 2 (CUSTOM_SIMPLE_EDGE) +Reducer 7 <- Reducer 2 (CUSTOM_SIMPLE_EDGE) +Reducer 9 <- Map 8 (CUSTOM_SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:100 + Stage-1 + Reducer 5 vectorized + File Output Operator [FS_250] + Limit [LIM_249] (rows=100 width=496) + Number of rows:100 + Select Operator [SEL_248] (rows=21091879 width=496) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6"] + <-Reducer 4 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_247] + Group By Operator [GBY_246] (rows=21091879 width=496) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6"],aggregations:["sum(VALUE._col0)","sum(VALUE._col1)","sum(VALUE._col2)"],keys:KEY._col0, KEY._col1, KEY._col2, KEY._col3 + <-Reducer 3 [SIMPLE_EDGE] + SHUFFLE [RS_47] + PartitionCols:_col0, _col1, _col2, _col3 + Group By Operator [GBY_46] (rows=21091879 width=496) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6"],aggregations:["sum(_col13)","sum(_col19)","sum(_col3)"],keys:_col6, _col7, _col22, _col23 + Top N Key Operator [TNK_91] (rows=4156223234 width=483) + keys:_col6, _col7, _col22, _col23,sort order:++++,top n:100 + Merge Join Operator [MERGEJOIN_203] (rows=4156223234 width=483) + Conds:RS_42._col2, _col1=RS_43._col11, _col12(Inner),Output:["_col3","_col6","_col7","_col13","_col19","_col22","_col23"] + <-Reducer 2 [SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_42] + PartitionCols:_col2, _col1 + Merge Join Operator [MERGEJOIN_197] (rows=7638375 width=10) + Conds:RS_214._col0=RS_206._col0(Inner),Output:["_col1","_col2","_col3"] + <-Map 8 [SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_206] + PartitionCols:_col0 + Select Operator [SEL_205] (rows=1957 width=4) + Output:["_col0"] + Filter Operator [FIL_204] (rows=1957 width=8) + predicate:(d_year) IN (1999, 2000, 2001) + TableScan [TS_3] (rows=73049 width=8) + default@date_dim,d3,Tbl:COMPLETE,Col:COMPLETE,Output:["d_date_sk","d_year"] + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_214] + PartitionCols:_col0 + Select Operator [SEL_213] (rows=285117831 width=15) + Output:["_col0","_col1","_col2","_col3"] + Filter Operator [FIL_212] (rows=285117831 width=15) + predicate:((cs_sold_date_sk BETWEEN DynamicValue(RS_40_d3_d_date_sk_min) AND DynamicValue(RS_40_d3_d_date_sk_max) and in_bloom_filter(cs_sold_date_sk, DynamicValue(RS_40_d3_d_date_sk_bloom_filter))) and cs_bill_customer_sk is not null and cs_sold_date_sk is not null) + TableScan [TS_0] (rows=287989836 width=15) + default@catalog_sales,catalog_sales,Tbl:COMPLETE,Col:COMPLETE,Output:["cs_sold_date_sk","cs_bill_customer_sk","cs_item_sk","cs_quantity"] + <-Reducer 9 [BROADCAST_EDGE] vectorized + BROADCAST [RS_211] + Group By Operator [GBY_210] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Map 8 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_209] + Group By Operator [GBY_208] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_207] (rows=1957 width=4) + Output:["_col0"] + Please refer to the previous Select Operator [SEL_205] + <-Reducer 14 [SIMPLE_EDGE] + SHUFFLE [RS_43] + PartitionCols:_col11, _col12 + Select Operator [SEL_38] (rows=21091879 width=484) + Output:["_col1","_col2","_col8","_col11","_col12","_col14","_col17","_col18"] + Merge Join Operator [MERGEJOIN_202] (rows=21091879 width=484) + Conds:RS_35._col3=RS_245._col0(Inner),Output:["_col5","_col8","_col9","_col11","_col14","_col15","_col17","_col18"] + <-Map 23 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_245] + PartitionCols:_col0 + Select Operator [SEL_244] (rows=1704 width=192) + Output:["_col0","_col1","_col2"] + TableScan [TS_24] (rows=1704 width=192) + default@store,store,Tbl:COMPLETE,Col:COMPLETE,Output:["s_store_sk","s_store_id","s_store_name"] + <-Reducer 13 [SIMPLE_EDGE] + SHUFFLE [RS_35] + PartitionCols:_col3 + Merge Join Operator [MERGEJOIN_201] (rows=21091879 width=298) + Conds:RS_32._col1=RS_243._col0(Inner),Output:["_col3","_col5","_col8","_col9","_col11","_col14","_col15"] + <-Map 22 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_243] + PartitionCols:_col0 + Select Operator [SEL_242] (rows=462000 width=288) + Output:["_col0","_col1","_col2"] + TableScan [TS_22] (rows=462000 width=288) + default@item,item,Tbl:COMPLETE,Col:COMPLETE,Output:["i_item_sk","i_item_id","i_item_desc"] + <-Reducer 12 [SIMPLE_EDGE] + SHUFFLE [RS_32] + PartitionCols:_col1 + Merge Join Operator [MERGEJOIN_200] (rows=21091879 width=18) + Conds:RS_29._col1, _col2, _col4=RS_30._col1, _col2, _col3(Inner),Output:["_col1","_col3","_col5","_col8","_col9","_col11"] + <-Reducer 17 [SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_30] + PartitionCols:_col1, _col2, _col3 + Merge Join Operator [MERGEJOIN_199] (rows=5384572 width=13) + Conds:RS_228._col0=RS_221._col0(Inner),Output:["_col1","_col2","_col3","_col4"] + <-Map 15 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_221] + PartitionCols:_col0 + Select Operator [SEL_218] (rows=201 width=4) + Output:["_col0"] + Filter Operator [FIL_216] (rows=201 width=12) + predicate:((d_year = 1999) and d_moy BETWEEN 4 AND 7) + TableScan [TS_9] (rows=73049 width=12) + default@date_dim,d1,Tbl:COMPLETE,Col:COMPLETE,Output:["d_date_sk","d_year","d_moy"] + <-Map 21 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_228] + PartitionCols:_col0 + Select Operator [SEL_227] (rows=53632139 width=19) + Output:["_col0","_col1","_col2","_col3","_col4"] + Filter Operator [FIL_226] (rows=53632139 width=19) + predicate:(sr_customer_sk is not null and sr_returned_date_sk is not null) + TableScan [TS_12] (rows=57591150 width=19) + default@store_returns,store_returns,Tbl:COMPLETE,Col:COMPLETE,Output:["sr_returned_date_sk","sr_item_sk","sr_customer_sk","sr_ticket_number","sr_return_quantity"] + <-Reducer 11 [SIMPLE_EDGE] + SHUFFLE [RS_29] + PartitionCols:_col1, _col2, _col4 + Merge Join Operator [MERGEJOIN_198] (rows=13737330 width=8) + Conds:RS_241._col0=RS_219._col0(Inner),Output:["_col1","_col2","_col3","_col4","_col5"] + <-Map 15 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_219] + PartitionCols:_col0 + Select Operator [SEL_217] (rows=50 width=4) + Output:["_col0"] + Filter Operator [FIL_215] (rows=50 width=12) + predicate:((d_moy = 4) and (d_year = 1999)) + Please refer to the previous TableScan [TS_9] + <-Map 10 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_241] + PartitionCols:_col0 + Select Operator [SEL_240] (rows=501694138 width=23) + Output:["_col0","_col1","_col2","_col3","_col4","_col5"] + Filter Operator [FIL_239] (rows=501694138 width=23) + predicate:((ss_customer_sk BETWEEN DynamicValue(RS_30_store_returns_sr_customer_sk_min) AND DynamicValue(RS_30_store_returns_sr_customer_sk_max) and in_bloom_filter(ss_customer_sk, DynamicValue(RS_30_store_returns_sr_customer_sk_bloom_filter))) and (ss_customer_sk BETWEEN DynamicValue(RS_42_catalog_sales_cs_bill_customer_sk_min) AND DynamicValue(RS_42_catalog_sales_cs_bill_customer_sk_max) and in_bloom_filter(ss_customer_sk, DynamicValue(RS_42_catalog_sales_cs_bill_customer_sk_bloom_filter))) and (ss_item_sk BETWEEN DynamicValue(RS_30_store_returns_sr_item_sk_min) AND DynamicValue(RS_30_store_returns_sr_item_sk_max) and in_bloom_filter(ss_item_sk, DynamicValue(RS_30_store_returns_sr_item_sk_bloom_filter))) and (ss_item_sk BETWEEN DynamicValue(RS_42_catalog_sales_cs_item_sk_min) AND DynamicValue(RS_42_catalog_sales_cs_item_sk_max) and in_bloom_filter(ss_item_sk, DynamicValue(RS_42_catalog_sales_cs_item_sk_bloom_filter))) and (ss_sold_da te_sk BETWEEN DynamicValue(RS_27_d1_d_date_sk_min) AND DynamicValue(RS_27_d1_d_date_sk_max) and in_bloom_filter(ss_sold_date_sk, DynamicValue(RS_27_d1_d_date_sk_bloom_filter))) and (ss_ticket_number BETWEEN DynamicValue(RS_30_store_returns_sr_ticket_number_min) AND DynamicValue(RS_30_store_returns_sr_ticket_number_max) and in_bloom_filter(ss_ticket_number, DynamicValue(RS_30_store_returns_sr_ticket_number_bloom_filter))) and ss_customer_sk is not null and ss_sold_date_sk is not null and ss_store_sk is not null) + TableScan [TS_6] (rows=575995635 width=23) + default@store_sales,store_sales,Tbl:COMPLETE,Col:COMPLETE,Output:["ss_sold_date_sk","ss_item_sk","ss_customer_sk","ss_store_sk","ss_ticket_number","ss_quantity"] + <-Reducer 16 [BROADCAST_EDGE] vectorized + BROADCAST [RS_225] + Group By Operator [GBY_224] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Map 15 [CUSTOM_SIMPLE_EDGE] vectorized + SHUFFLE [RS_223] + Group By Operator [GBY_222] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_220] (rows=50 width=4) + Output:["_col0"] + Please refer to the previous Select Operator [SEL_217] + <-Reducer 18 [BROADCAST_EDGE] vectorized + BROADCAST [RS_230] + Group By Operator [GBY_229] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Reducer 17 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_119] + Group By Operator [GBY_118] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_117] (rows=5384572 width=8) + Output:["_col0"] + Please refer to the previous Merge Join Operator [MERGEJOIN_199] + <-Reducer 19 [BROADCAST_EDGE] vectorized + BROADCAST [RS_232] + Group By Operator [GBY_231] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Reducer 17 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_124] + Group By Operator [GBY_123] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_122] (rows=5384572 width=5) + Output:["_col0"] + Please refer to the previous Merge Join Operator [MERGEJOIN_199] + <-Reducer 20 [BROADCAST_EDGE] vectorized + BROADCAST [RS_234] + Group By Operator [GBY_233] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=3507020)"] + <-Reducer 17 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_129] + Group By Operator [GBY_128] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=3507020)"] + Select Operator [SEL_127] (rows=5384572 width=8) + Output:["_col0"] + Please refer to the previous Merge Join Operator [MERGEJOIN_199] + <-Reducer 6 [BROADCAST_EDGE] vectorized + BROADCAST [RS_236] + Group By Operator [GBY_235] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Reducer 2 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_144] + Group By Operator [GBY_143] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_142] (rows=7638375 width=8) + Output:["_col0"] + Please refer to the previous Merge Join Operator [MERGEJOIN_197] + <-Reducer 7 [BROADCAST_EDGE] vectorized + BROADCAST [RS_238] + Group By Operator [GBY_237] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Reducer 2 [CUSTOM_SIMPLE_EDGE] + PARTITION_ONLY_SHUFFLE [RS_149] + Group By Operator [GBY_148] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_147] (rows=7638375 width=6) + Output:["_col0"] + Please refer to the previous Merge Join Operator [MERGEJOIN_197] + http://git-wip-us.apache.org/repos/asf/hive/blob/b8299551/ql/src/test/results/clientpositive/perf/tez/constraints/query3.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/tez/constraints/query3.q.out b/ql/src/test/results/clientpositive/perf/tez/constraints/query3.q.out new file mode 100644 index 0000000..0cca24c --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/tez/constraints/query3.q.out @@ -0,0 +1,135 @@ +PREHOOK: query: explain +select dt.d_year + ,item.i_brand_id brand_id + ,item.i_brand brand + ,sum(ss_ext_sales_price) sum_agg + from date_dim dt + ,store_sales + ,item + where dt.d_date_sk = store_sales.ss_sold_date_sk + and store_sales.ss_item_sk = item.i_item_sk + and item.i_manufact_id = 436 + and dt.d_moy=12 + group by dt.d_year + ,item.i_brand + ,item.i_brand_id + order by dt.d_year + ,sum_agg desc + ,brand_id + limit 100 +PREHOOK: type: QUERY +PREHOOK: Input: default@date_dim +PREHOOK: Input: default@item +PREHOOK: Input: default@store_sales +PREHOOK: Output: hdfs://### HDFS PATH ### +POSTHOOK: query: explain +select dt.d_year + ,item.i_brand_id brand_id + ,item.i_brand brand + ,sum(ss_ext_sales_price) sum_agg + from date_dim dt + ,store_sales + ,item + where dt.d_date_sk = store_sales.ss_sold_date_sk + and store_sales.ss_item_sk = item.i_item_sk + and item.i_manufact_id = 436 + and dt.d_moy=12 + group by dt.d_year + ,item.i_brand + ,item.i_brand_id + order by dt.d_year + ,sum_agg desc + ,brand_id + limit 100 +POSTHOOK: type: QUERY +POSTHOOK: Input: default@date_dim +POSTHOOK: Input: default@item +POSTHOOK: Input: default@store_sales +POSTHOOK: Output: hdfs://### HDFS PATH ### +Plan optimized by CBO. + +Vertex dependency in root stage +Map 1 <- Reducer 7 (BROADCAST_EDGE), Reducer 9 (BROADCAST_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 6 (SIMPLE_EDGE) +Reducer 3 <- Map 8 (SIMPLE_EDGE), Reducer 2 (SIMPLE_EDGE) +Reducer 4 <- Reducer 3 (SIMPLE_EDGE) +Reducer 5 <- Reducer 4 (SIMPLE_EDGE) +Reducer 7 <- Map 6 (CUSTOM_SIMPLE_EDGE) +Reducer 9 <- Map 8 (CUSTOM_SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:100 + Stage-1 + Reducer 5 vectorized + File Output Operator [FS_77] + Limit [LIM_76] (rows=100 width=220) + Number of rows:100 + Select Operator [SEL_75] (rows=274400 width=220) + Output:["_col0","_col1","_col2","_col3"] + <-Reducer 4 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_74] + Group By Operator [GBY_73] (rows=274400 width=220) + Output:["_col0","_col1","_col2","_col3"],aggregations:["sum(VALUE._col0)"],keys:KEY._col0, KEY._col1, KEY._col2 + <-Reducer 3 [SIMPLE_EDGE] + SHUFFLE [RS_17] + PartitionCols:_col0, _col1, _col2 + Group By Operator [GBY_16] (rows=274400 width=220) + Output:["_col0","_col1","_col2","_col3"],aggregations:["sum(_col2)"],keys:_col7, _col4, _col5 + Merge Join Operator [MERGEJOIN_53] (rows=589741 width=108) + Conds:RS_12._col0=RS_64._col0(Inner),Output:["_col2","_col4","_col5","_col7"] + <-Map 8 [SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_64] + PartitionCols:_col0 + Select Operator [SEL_63] (rows=5619 width=8) + Output:["_col0","_col1"] + Filter Operator [FIL_62] (rows=5619 width=12) + predicate:(d_moy = 12) + TableScan [TS_6] (rows=73049 width=12) + default@date_dim,dt,Tbl:COMPLETE,Col:COMPLETE,Output:["d_date_sk","d_year","d_moy"] + <-Reducer 2 [SIMPLE_EDGE] + SHUFFLE [RS_12] + PartitionCols:_col0 + Merge Join Operator [MERGEJOIN_52] (rows=7666836 width=104) + Conds:RS_72._col1=RS_56._col0(Inner),Output:["_col0","_col2","_col4","_col5"] + <-Map 6 [SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_56] + PartitionCols:_col0 + Select Operator [SEL_55] (rows=669 width=107) + Output:["_col0","_col1","_col2"] + Filter Operator [FIL_54] (rows=669 width=111) + predicate:(i_manufact_id = 436) + TableScan [TS_3] (rows=462000 width=111) + default@item,item,Tbl:COMPLETE,Col:COMPLETE,Output:["i_item_sk","i_brand_id","i_brand","i_manufact_id"] + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_72] + PartitionCols:_col1 + Select Operator [SEL_71] (rows=550076554 width=114) + Output:["_col0","_col1","_col2"] + Filter Operator [FIL_70] (rows=550076554 width=114) + predicate:((ss_item_sk BETWEEN DynamicValue(RS_10_item_i_item_sk_min) AND DynamicValue(RS_10_item_i_item_sk_max) and in_bloom_filter(ss_item_sk, DynamicValue(RS_10_item_i_item_sk_bloom_filter))) and (ss_sold_date_sk BETWEEN DynamicValue(RS_13_dt_d_date_sk_min) AND DynamicValue(RS_13_dt_d_date_sk_max) and in_bloom_filter(ss_sold_date_sk, DynamicValue(RS_13_dt_d_date_sk_bloom_filter))) and ss_sold_date_sk is not null) + TableScan [TS_0] (rows=575995635 width=114) + default@store_sales,store_sales,Tbl:COMPLETE,Col:COMPLETE,Output:["ss_sold_date_sk","ss_item_sk","ss_ext_sales_price"] + <-Reducer 7 [BROADCAST_EDGE] vectorized + BROADCAST [RS_61] + Group By Operator [GBY_60] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Map 6 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_59] + Group By Operator [GBY_58] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_57] (rows=669 width=4) + Output:["_col0"] + Please refer to the previous Select Operator [SEL_55] + <-Reducer 9 [BROADCAST_EDGE] vectorized + BROADCAST [RS_69] + Group By Operator [GBY_68] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(VALUE._col0)","max(VALUE._col1)","bloom_filter(VALUE._col2, expectedEntries=1000000)"] + <-Map 8 [CUSTOM_SIMPLE_EDGE] vectorized + PARTITION_ONLY_SHUFFLE [RS_67] + Group By Operator [GBY_66] (rows=1 width=12) + Output:["_col0","_col1","_col2"],aggregations:["min(_col0)","max(_col0)","bloom_filter(_col0, expectedEntries=1000000)"] + Select Operator [SEL_65] (rows=5619 width=4) + Output:["_col0"] + Please refer to the previous Select Operator [SEL_63] + http://git-wip-us.apache.org/repos/asf/hive/blob/b8299551/ql/src/test/results/clientpositive/perf/tez/constraints/query30.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/tez/constraints/query30.q.out b/ql/src/test/results/clientpositive/perf/tez/constraints/query30.q.out new file mode 100644 index 0000000..7924edd --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/tez/constraints/query30.q.out @@ -0,0 +1,221 @@ +PREHOOK: query: explain +with customer_total_return as + (select wr_returning_customer_sk as ctr_customer_sk + ,ca_state as ctr_state, + sum(wr_return_amt) as ctr_total_return + from web_returns + ,date_dim + ,customer_address + where wr_returned_date_sk = d_date_sk + and d_year =2002 + and wr_returning_addr_sk = ca_address_sk + group by wr_returning_customer_sk + ,ca_state) + select c_customer_id,c_salutation,c_first_name,c_last_name,c_preferred_cust_flag + ,c_birth_day,c_birth_month,c_birth_year,c_birth_country,c_login,c_email_address + ,c_last_review_date,ctr_total_return + from customer_total_return ctr1 + ,customer_address + ,customer + where ctr1.ctr_total_return > (select avg(ctr_total_return)*1.2 + from customer_total_return ctr2 + where ctr1.ctr_state = ctr2.ctr_state) + and ca_address_sk = c_current_addr_sk + and ca_state = 'IL' + and ctr1.ctr_customer_sk = c_customer_sk + order by c_customer_id,c_salutation,c_first_name,c_last_name,c_preferred_cust_flag + ,c_birth_day,c_birth_month,c_birth_year,c_birth_country,c_login,c_email_address + ,c_last_review_date,ctr_total_return +limit 100 +PREHOOK: type: QUERY +PREHOOK: Input: default@customer +PREHOOK: Input: default@customer_address +PREHOOK: Input: default@date_dim +PREHOOK: Input: default@web_returns +PREHOOK: Output: hdfs://### HDFS PATH ### +POSTHOOK: query: explain +with customer_total_return as + (select wr_returning_customer_sk as ctr_customer_sk + ,ca_state as ctr_state, + sum(wr_return_amt) as ctr_total_return + from web_returns + ,date_dim + ,customer_address + where wr_returned_date_sk = d_date_sk + and d_year =2002 + and wr_returning_addr_sk = ca_address_sk + group by wr_returning_customer_sk + ,ca_state) + select c_customer_id,c_salutation,c_first_name,c_last_name,c_preferred_cust_flag + ,c_birth_day,c_birth_month,c_birth_year,c_birth_country,c_login,c_email_address + ,c_last_review_date,ctr_total_return + from customer_total_return ctr1 + ,customer_address + ,customer + where ctr1.ctr_total_return > (select avg(ctr_total_return)*1.2 + from customer_total_return ctr2 + where ctr1.ctr_state = ctr2.ctr_state) + and ca_address_sk = c_current_addr_sk + and ca_state = 'IL' + and ctr1.ctr_customer_sk = c_customer_sk + order by c_customer_id,c_salutation,c_first_name,c_last_name,c_preferred_cust_flag + ,c_birth_day,c_birth_month,c_birth_year,c_birth_country,c_login,c_email_address + ,c_last_review_date,ctr_total_return +limit 100 +POSTHOOK: type: QUERY +POSTHOOK: Input: default@customer +POSTHOOK: Input: default@customer_address +POSTHOOK: Input: default@date_dim +POSTHOOK: Input: default@web_returns +POSTHOOK: Output: hdfs://### HDFS PATH ### +Plan optimized by CBO. + +Vertex dependency in root stage +Reducer 10 <- Reducer 9 (SIMPLE_EDGE) +Reducer 12 <- Map 11 (SIMPLE_EDGE), Map 14 (SIMPLE_EDGE) +Reducer 13 <- Map 11 (SIMPLE_EDGE), Map 14 (SIMPLE_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 5 (SIMPLE_EDGE) +Reducer 3 <- Reducer 2 (SIMPLE_EDGE), Reducer 8 (SIMPLE_EDGE) +Reducer 4 <- Reducer 3 (SIMPLE_EDGE) +Reducer 6 <- Map 5 (SIMPLE_EDGE), Reducer 12 (SIMPLE_EDGE) +Reducer 7 <- Reducer 6 (SIMPLE_EDGE) +Reducer 8 <- Reducer 10 (SIMPLE_EDGE), Reducer 7 (SIMPLE_EDGE) +Reducer 9 <- Map 5 (SIMPLE_EDGE), Reducer 13 (SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:100 + Stage-1 + Reducer 4 vectorized + File Output Operator [FS_210] + Limit [LIM_209] (rows=100 width=942) + Number of rows:100 + Select Operator [SEL_208] (rows=691171 width=942) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12"] + <-Reducer 3 [SIMPLE_EDGE] + SHUFFLE [RS_63] + Select Operator [SEL_62] (rows=691171 width=942) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12"] + Merge Join Operator [MERGEJOIN_177] (rows=691171 width=942) + Conds:RS_59._col0=RS_60._col0(Inner),Output:["_col1","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12","_col13","_col17"] + <-Reducer 2 [SIMPLE_EDGE] + SHUFFLE [RS_59] + PartitionCols:_col0 + Merge Join Operator [MERGEJOIN_171] (rows=1568628 width=834) + Conds:RS_180._col2=RS_187._col0(Inner),Output:["_col0","_col1","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12","_col13"] + <-Map 5 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_187] + PartitionCols:_col0 + Select Operator [SEL_184] (rows=784314 width=4) + Output:["_col0"] + Filter Operator [FIL_181] (rows=784314 width=90) + predicate:(ca_state = 'IL') + TableScan [TS_3] (rows=40000000 width=90) + default@customer_address,customer_address,Tbl:COMPLETE,Col:COMPLETE,Output:["ca_address_sk","ca_state"] + <-Map 1 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_180] + PartitionCols:_col2 + Select Operator [SEL_179] (rows=80000000 width=849) + Output:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7","_col8","_col9","_col10","_col11","_col12","_col13"] + Filter Operator [FIL_178] (rows=80000000 width=849) + predicate:c_current_addr_sk is not null + TableScan [TS_0] (rows=80000000 width=849) + default@customer,customer,Tbl:COMPLETE,Col:COMPLETE,Output:["c_customer_sk","c_customer_id","c_current_addr_sk","c_salutation","c_first_name","c_last_name","c_preferred_cust_flag","c_birth_day","c_birth_month","c_birth_year","c_birth_country","c_login","c_email_address","c_last_review_date"] + <-Reducer 8 [SIMPLE_EDGE] + SHUFFLE [RS_60] + PartitionCols:_col0 + Select Operator [SEL_55] (rows=704993 width=227) + Output:["_col0","_col2"] + Filter Operator [FIL_54] (rows=704993 width=227) + predicate:(_col2 > _col3) + Merge Join Operator [MERGEJOIN_176] (rows=2114980 width=227) + Conds:RS_202._col1=RS_207._col1(Inner),Output:["_col0","_col2","_col3"] + <-Reducer 10 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_207] + PartitionCols:_col1 + Select Operator [SEL_206] (rows=6 width=198) + Output:["_col0","_col1"] + Group By Operator [GBY_205] (rows=6 width=206) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col2)","count(_col2)"],keys:_col0 + Select Operator [SEL_204] (rows=2537976 width=201) + Output:["_col0","_col2"] + Group By Operator [GBY_203] (rows=2537976 width=201) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)"],keys:KEY._col0, KEY._col1 + <-Reducer 9 [SIMPLE_EDGE] + SHUFFLE [RS_43] + PartitionCols:_col0 + Group By Operator [GBY_42] (rows=3923529 width=201) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col3)"],keys:_col6, _col1 + Merge Join Operator [MERGEJOIN_175] (rows=3923529 width=184) + Conds:RS_38._col2=RS_189._col0(Inner),Output:["_col1","_col3","_col6"] + <-Map 5 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_189] + PartitionCols:_col0 + Select Operator [SEL_186] (rows=40000000 width=90) + Output:["_col0","_col1"] + Filter Operator [FIL_183] (rows=40000000 width=90) + predicate:ca_state is not null + Please refer to the previous TableScan [TS_3] + <-Reducer 13 [SIMPLE_EDGE] + SHUFFLE [RS_38] + PartitionCols:_col2 + Merge Join Operator [MERGEJOIN_174] (rows=3923529 width=101) + Conds:RS_195._col0=RS_199._col0(Inner),Output:["_col1","_col2","_col3"] + <-Map 11 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_195] + PartitionCols:_col0 + Select Operator [SEL_193] (rows=13130761 width=118) + Output:["_col0","_col1","_col2","_col3"] + Filter Operator [FIL_191] (rows=13130761 width=118) + predicate:(wr_returned_date_sk is not null and wr_returning_addr_sk is not null) + TableScan [TS_6] (rows=14398467 width=118) + default@web_returns,web_returns,Tbl:COMPLETE,Col:COMPLETE,Output:["wr_returned_date_sk","wr_returning_customer_sk","wr_returning_addr_sk","wr_return_amt"] + <-Map 14 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_199] + PartitionCols:_col0 + Select Operator [SEL_197] (rows=652 width=4) + Output:["_col0"] + Filter Operator [FIL_196] (rows=652 width=8) + predicate:(d_year = 2002) + TableScan [TS_9] (rows=73049 width=8) + default@date_dim,date_dim,Tbl:COMPLETE,Col:COMPLETE,Output:["d_date_sk","d_year"] + <-Reducer 7 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_202] + PartitionCols:_col1 + Select Operator [SEL_201] (rows=2114980 width=201) + Output:["_col0","_col1","_col2"] + Group By Operator [GBY_200] (rows=2114980 width=201) + Output:["_col0","_col1","_col2"],aggregations:["sum(VALUE._col0)"],keys:KEY._col0, KEY._col1 + <-Reducer 6 [SIMPLE_EDGE] + SHUFFLE [RS_23] + PartitionCols:_col0, _col1 + Group By Operator [GBY_22] (rows=3746772 width=201) + Output:["_col0","_col1","_col2"],aggregations:["sum(_col3)"],keys:_col6, _col1 + Merge Join Operator [MERGEJOIN_173] (rows=3746772 width=184) + Conds:RS_18._col2=RS_188._col0(Inner),Output:["_col1","_col3","_col6"] + <-Map 5 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_188] + PartitionCols:_col0 + Select Operator [SEL_185] (rows=40000000 width=90) + Output:["_col0","_col1"] + Filter Operator [FIL_182] (rows=40000000 width=90) + predicate:ca_state is not null + Please refer to the previous TableScan [TS_3] + <-Reducer 12 [SIMPLE_EDGE] + SHUFFLE [RS_18] + PartitionCols:_col2 + Merge Join Operator [MERGEJOIN_172] (rows=3746772 width=101) + Conds:RS_194._col0=RS_198._col0(Inner),Output:["_col1","_col2","_col3"] + <-Map 11 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_194] + PartitionCols:_col0 + Select Operator [SEL_192] (rows=12539215 width=118) + Output:["_col0","_col1","_col2","_col3"] + Filter Operator [FIL_190] (rows=12539215 width=118) + predicate:(wr_returned_date_sk is not null and wr_returning_addr_sk is not null and wr_returning_customer_sk is not null) + Please refer to the previous TableScan [TS_6] + <-Map 14 [SIMPLE_EDGE] vectorized + SHUFFLE [RS_198] + PartitionCols:_col0 + Please refer to the previous Select Operator [SEL_197] +
