http://git-wip-us.apache.org/repos/asf/hive/blob/09b6f9a3/ql/src/test/results/clientpositive/perf/query84.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/query84.q.out b/ql/src/test/results/clientpositive/perf/query84.q.out new file mode 100644 index 0000000..8e15aee --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/query84.q.out @@ -0,0 +1,171 @@ +PREHOOK: query: explain select c_customer_id as customer_id ,concat(c_last_name, ', ', c_first_name) as customername from customer ,customer_address ,customer_demographics ,household_demographics ,income_band ,store_returns where ca_city = 'Hopewell' and customer.c_current_addr_sk = customer_address.ca_address_sk and ib_lower_bound >= 32287 and ib_upper_bound <= 32287 + 50000 and income_band.ib_income_band_sk = household_demographics.hd_income_band_sk and customer_demographics.cd_demo_sk = customer.c_current_cdemo_sk and household_demographics.hd_demo_sk = customer.c_current_hdemo_sk and store_returns.sr_cdemo_sk = customer_demographics.cd_demo_sk order by customer_id limit 100 +PREHOOK: type: QUERY +POSTHOOK: query: explain select c_customer_id as customer_id ,concat(c_last_name, ', ', c_first_name) as customername from customer ,customer_address ,customer_demographics ,household_demographics ,income_band ,store_returns where ca_city = 'Hopewell' and customer.c_current_addr_sk = customer_address.ca_address_sk and ib_lower_bound >= 32287 and ib_upper_bound <= 32287 + 50000 and income_band.ib_income_band_sk = household_demographics.hd_income_band_sk and customer_demographics.cd_demo_sk = customer.c_current_cdemo_sk and household_demographics.hd_demo_sk = customer.c_current_hdemo_sk and store_returns.sr_cdemo_sk = customer_demographics.cd_demo_sk order by customer_id limit 100 +POSTHOOK: type: QUERY +Plan optimized by CBO. + +Vertex dependency in root stage +Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 7 (SIMPLE_EDGE) +Reducer 3 <- Map 11 (SIMPLE_EDGE), Map 8 (SIMPLE_EDGE), Reducer 2 (SIMPLE_EDGE) +Reducer 4 <- Map 9 (SIMPLE_EDGE), Reducer 3 (SIMPLE_EDGE) +Reducer 5 <- Map 10 (SIMPLE_EDGE), Reducer 4 (SIMPLE_EDGE) +Reducer 6 <- Reducer 5 (SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:100 + Stage-1 + Reducer 6 + File Output Operator [FS_44] + compressed:false + Statistics:Num rows: 100 Data size: 86000 Basic stats: COMPLETE Column stats: NONE + table:{"input format:":"org.apache.hadoop.mapred.TextInputFormat","output format:":"org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat","serde:":"org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe"} + Limit [LIM_43] + Number of rows:100 + Statistics:Num rows: 100 Data size: 86000 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_42] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 234256017 Data size: 201464909002 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 5 [SIMPLE_EDGE] + Reduce Output Operator [RS_41] + key expressions:_col0 (type: string) + sort order:+ + Statistics:Num rows: 234256017 Data size: 201464909002 Basic stats: COMPLETE Column stats: NONE + value expressions:_col1 (type: string) + Select Operator [SEL_40] + outputColumnNames:["_col0","_col1"] + Statistics:Num rows: 234256017 Data size: 201464909002 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_75] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col11 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col4","_col5"] + | Statistics:Num rows: 234256017 Data size: 201464909002 Basic stats: COMPLETE Column stats: NONE + |<-Map 10 [SIMPLE_EDGE] + | Reduce Output Operator [RS_38] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 2 Data size: 24 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_14] + | outputColumnNames:["_col0"] + | Statistics:Num rows: 2 Data size: 24 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_70] + | predicate:(((ib_upper_bound <= 82287) and (ib_lower_bound >= 32287)) and ib_income_band_sk is not null) (type: boolean) + | Statistics:Num rows: 2 Data size: 24 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_12] + | alias:income_band + | Statistics:Num rows: 20 Data size: 240 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 4 [SIMPLE_EDGE] + Reduce Output Operator [RS_36] + key expressions:_col11 (type: int) + Map-reduce partition columns:_col11 (type: int) + sort order:+ + Statistics:Num rows: 212960011 Data size: 183149913305 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: string), _col4 (type: string), _col5 (type: string) + Merge Join Operator [MERGEJOIN_74] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col2 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col4","_col5","_col11"] + | Statistics:Num rows: 212960011 Data size: 183149913305 Basic stats: COMPLETE Column stats: NONE + |<-Map 9 [SIMPLE_EDGE] + | Reduce Output Operator [RS_33] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 7200 Data size: 770400 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: int) + | Select Operator [SEL_11] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 7200 Data size: 770400 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_69] + | predicate:(hd_demo_sk is not null and hd_income_band_sk is not null) (type: boolean) + | Statistics:Num rows: 7200 Data size: 770400 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_9] + | alias:household_demographics + | Statistics:Num rows: 7200 Data size: 770400 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 3 [SIMPLE_EDGE] + Reduce Output Operator [RS_31] + key expressions:_col2 (type: int) + Map-reduce partition columns:_col2 (type: int) + sort order:+ + Statistics:Num rows: 193600006 Data size: 166499917578 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: string), _col4 (type: string), _col5 (type: string) + Merge Join Operator [MERGEJOIN_73] + | condition map:[{"":"Inner Join 0 to 1"},{"":"Inner Join 1 to 2"}] + | keys:{"0":"_col1 (type: int)","1":"_col0 (type: int)","2":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col2","_col4","_col5"] + | Statistics:Num rows: 193600006 Data size: 166499917578 Basic stats: COMPLETE Column stats: NONE + |<-Map 11 [SIMPLE_EDGE] + | Reduce Output Operator [RS_28] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | Select Operator [SEL_17] + | outputColumnNames:["_col0"] + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | Filter Operator [FIL_71] + | predicate:sr_cdemo_sk is not null (type: boolean) + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | TableScan [TS_15] + | alias:store_returns + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + |<-Map 8 [SIMPLE_EDGE] + | Reduce Output Operator [RS_26] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_8] + | outputColumnNames:["_col0"] + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_68] + | predicate:cd_demo_sk is not null (type: boolean) + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_6] + | alias:customer_demographics + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 2 [SIMPLE_EDGE] + Reduce Output Operator [RS_24] + key expressions:_col1 (type: int) + Map-reduce partition columns:_col1 (type: int) + sort order:+ + Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: string), _col2 (type: int), _col4 (type: string), _col5 (type: string) + Merge Join Operator [MERGEJOIN_72] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col3 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col1","_col2","_col4","_col5"] + | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + |<-Map 1 [SIMPLE_EDGE] + | Reduce Output Operator [RS_19] + | key expressions:_col3 (type: int) + | Map-reduce partition columns:_col3 (type: int) + | sort order:+ + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col0 (type: string), _col1 (type: int), _col2 (type: int), _col4 (type: string), _col5 (type: string) + | Select Operator [SEL_2] + | outputColumnNames:["_col0","_col1","_col2","_col3","_col4","_col5"] + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_66] + | predicate:((c_current_addr_sk is not null and c_current_cdemo_sk is not null) and c_current_hdemo_sk is not null) (type: boolean) + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_0] + | alias:customer + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + |<-Map 7 [SIMPLE_EDGE] + Reduce Output Operator [RS_21] + key expressions:_col0 (type: int) + Map-reduce partition columns:_col0 (type: int) + sort order:+ + Statistics:Num rows: 20000000 Data size: 20297597642 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_5] + outputColumnNames:["_col0"] + Statistics:Num rows: 20000000 Data size: 20297597642 Basic stats: COMPLETE Column stats: NONE + Filter Operator [FIL_67] + predicate:((ca_city = 'Hopewell') and ca_address_sk is not null) (type: boolean) + Statistics:Num rows: 20000000 Data size: 20297597642 Basic stats: COMPLETE Column stats: NONE + TableScan [TS_3] + alias:customer_address + Statistics:Num rows: 40000000 Data size: 40595195284 Basic stats: COMPLETE Column stats: NONE +
http://git-wip-us.apache.org/repos/asf/hive/blob/09b6f9a3/ql/src/test/results/clientpositive/perf/query85.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/query85.q.out b/ql/src/test/results/clientpositive/perf/query85.q.out new file mode 100644 index 0000000..660c578 --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/query85.q.out @@ -0,0 +1,275 @@ +PREHOOK: query: explain select substr(r_reason_desc,1,20) as r ,avg(ws_quantity) wq ,avg(wr_refunded_cash) ref ,avg(wr_fee) fee from web_sales, web_returns, web_page, customer_demographics cd1, customer_demographics cd2, customer_address, date_dim, reason where web_sales.ws_web_page_sk = web_page.wp_web_page_sk and web_sales.ws_item_sk = web_returns.wr_item_sk and web_sales.ws_order_number = web_returns.wr_order_number and web_sales.ws_sold_date_sk = date_dim.d_date_sk and d_year = 1998 and cd1.cd_demo_sk = web_returns.wr_refunded_cdemo_sk and cd2.cd_demo_sk = web_returns.wr_returning_cdemo_sk and customer_address.ca_address_sk = web_returns.wr_refunded_addr_sk and reason.r_reason_sk = web_returns.wr_reason_sk and ( ( cd1.cd_marital_status = 'M' and cd1.cd_marital_status = cd2.cd_marital_status and cd1.cd_education_status = '4 yr Degree' and cd1.cd_education_status = cd2.cd_education_status and ws_sales_price between 100.00 and 150.00 ) or ( cd1.cd_marital_status = 'D' and cd1.cd_ma rital_status = cd2.cd_marital_status and cd1.cd_education_status = 'Primary' and cd1.cd_education_status = cd2.cd_education_status and ws_sales_price between 50.00 and 100.00 ) or ( cd1.cd_marital_status = 'U' and cd1.cd_marital_status = cd2.cd_marital_status and cd1.cd_education_status = 'Advanced Degree' and cd1.cd_education_status = cd2.cd_education_status and ws_sales_price between 150.00 and 200.00 ) ) and ( ( ca_country = 'United States' and ca_state in ('KY', 'GA', 'NM') and ws_net_profit between 100 and 200 ) or ( ca_country = 'United States' and ca_state in ('MT', 'OR', 'IN') and ws_net_profit between 150 and 300 ) or ( ca_country = 'United States' and ca_state in ('WI', 'MO', 'WV') and ws_net_profit between 50 and 250 ) ) group by r_reason_desc order by r, wq, ref, fee limit 100 +PREHOOK: type: QUERY +POSTHOOK: query: explain select substr(r_reason_desc,1,20) as r ,avg(ws_quantity) wq ,avg(wr_refunded_cash) ref ,avg(wr_fee) fee from web_sales, web_returns, web_page, customer_demographics cd1, customer_demographics cd2, customer_address, date_dim, reason where web_sales.ws_web_page_sk = web_page.wp_web_page_sk and web_sales.ws_item_sk = web_returns.wr_item_sk and web_sales.ws_order_number = web_returns.wr_order_number and web_sales.ws_sold_date_sk = date_dim.d_date_sk and d_year = 1998 and cd1.cd_demo_sk = web_returns.wr_refunded_cdemo_sk and cd2.cd_demo_sk = web_returns.wr_returning_cdemo_sk and customer_address.ca_address_sk = web_returns.wr_refunded_addr_sk and reason.r_reason_sk = web_returns.wr_reason_sk and ( ( cd1.cd_marital_status = 'M' and cd1.cd_marital_status = cd2.cd_marital_status and cd1.cd_education_status = '4 yr Degree' and cd1.cd_education_status = cd2.cd_education_status and ws_sales_price between 100.00 and 150.00 ) or ( cd1.cd_marital_status = 'D' and cd1.cd_m arital_status = cd2.cd_marital_status and cd1.cd_education_status = 'Primary' and cd1.cd_education_status = cd2.cd_education_status and ws_sales_price between 50.00 and 100.00 ) or ( cd1.cd_marital_status = 'U' and cd1.cd_marital_status = cd2.cd_marital_status and cd1.cd_education_status = 'Advanced Degree' and cd1.cd_education_status = cd2.cd_education_status and ws_sales_price between 150.00 and 200.00 ) ) and ( ( ca_country = 'United States' and ca_state in ('KY', 'GA', 'NM') and ws_net_profit between 100 and 200 ) or ( ca_country = 'United States' and ca_state in ('MT', 'OR', 'IN') and ws_net_profit between 150 and 300 ) or ( ca_country = 'United States' and ca_state in ('WI', 'MO', 'WV') and ws_net_profit between 50 and 250 ) ) group by r_reason_desc order by r, wq, ref, fee limit 100 +POSTHOOK: type: QUERY +Plan optimized by CBO. + +Vertex dependency in root stage +Reducer 10 <- Reducer 9 (SIMPLE_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 11 (SIMPLE_EDGE) +Reducer 3 <- Map 12 (SIMPLE_EDGE), Reducer 2 (SIMPLE_EDGE) +Reducer 4 <- Map 13 (SIMPLE_EDGE), Reducer 3 (SIMPLE_EDGE) +Reducer 5 <- Map 14 (SIMPLE_EDGE), Reducer 4 (SIMPLE_EDGE) +Reducer 6 <- Map 15 (SIMPLE_EDGE), Reducer 5 (SIMPLE_EDGE) +Reducer 7 <- Map 16 (SIMPLE_EDGE), Reducer 6 (SIMPLE_EDGE) +Reducer 8 <- Map 17 (SIMPLE_EDGE), Reducer 7 (SIMPLE_EDGE) +Reducer 9 <- Reducer 8 (SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:100 + Stage-1 + Reducer 10 + File Output Operator [FS_71] + compressed:false + Statistics:Num rows: 100 Data size: 101400 Basic stats: COMPLETE Column stats: NONE + table:{"input format:":"org.apache.hadoop.mapred.TextInputFormat","output format:":"org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat","serde:":"org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe"} + Limit [LIM_70] + Number of rows:100 + Statistics:Num rows: 100 Data size: 101400 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_69] + | outputColumnNames:["_col0","_col1","_col2","_col3"] + | Statistics:Num rows: 9982500 Data size: 10131039080 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 9 [SIMPLE_EDGE] + Reduce Output Operator [RS_68] + key expressions:_col0 (type: string), _col1 (type: double), _col2 (type: decimal(11,6)), _col3 (type: decimal(11,6)) + sort order:++++ + Statistics:Num rows: 9982500 Data size: 10131039080 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_67] + outputColumnNames:["_col0","_col1","_col2","_col3"] + Statistics:Num rows: 9982500 Data size: 10131039080 Basic stats: COMPLETE Column stats: NONE + Group By Operator [GBY_66] + | aggregations:["avg(VALUE._col0)","avg(VALUE._col1)","avg(VALUE._col2)"] + | keys:KEY._col0 (type: string) + | outputColumnNames:["_col0","_col1","_col2","_col3"] + | Statistics:Num rows: 9982500 Data size: 10131039080 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 8 [SIMPLE_EDGE] + Reduce Output Operator [RS_65] + key expressions:_col0 (type: string) + Map-reduce partition columns:_col0 (type: string) + sort order:+ + Statistics:Num rows: 19965000 Data size: 20262078161 Basic stats: COMPLETE Column stats: NONE + value expressions:_col1 (type: struct<count:bigint,sum:double,input:int>), _col2 (type: struct<count:bigint,sum:decimal(17,2),input:decimal(7,2)>), _col3 (type: struct<count:bigint,sum:decimal(17,2),input:decimal(7,2)>) + Group By Operator [GBY_64] + aggregations:["avg(_col4)","avg(_col14)","avg(_col13)"] + keys:_col28 (type: string) + outputColumnNames:["_col0","_col1","_col2","_col3"] + Statistics:Num rows: 19965000 Data size: 20262078161 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_63] + outputColumnNames:["_col28","_col4","_col14","_col13"] + Statistics:Num rows: 19965000 Data size: 20262078161 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_123] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col11 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col4","_col13","_col14","_col28"] + | Statistics:Num rows: 19965000 Data size: 20262078161 Basic stats: COMPLETE Column stats: NONE + |<-Map 17 [SIMPLE_EDGE] + | Reduce Output Operator [RS_61] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 72 Data size: 14400 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: string) + | Select Operator [SEL_52] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 72 Data size: 14400 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_116] + | predicate:r_reason_sk is not null (type: boolean) + | Statistics:Num rows: 72 Data size: 14400 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_50] + | alias:reason + | Statistics:Num rows: 72 Data size: 14400 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 7 [SIMPLE_EDGE] + Reduce Output Operator [RS_59] + key expressions:_col11 (type: int) + Map-reduce partition columns:_col11 (type: int) + sort order:+ + Statistics:Num rows: 18150000 Data size: 18420070657 Basic stats: COMPLETE Column stats: NONE + value expressions:_col4 (type: int), _col13 (type: decimal(7,2)), _col14 (type: decimal(7,2)) + Merge Join Operator [MERGEJOIN_122] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col0 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col4","_col11","_col13","_col14"] + | Statistics:Num rows: 18150000 Data size: 18420070657 Basic stats: COMPLETE Column stats: NONE + |<-Map 16 [SIMPLE_EDGE] + | Reduce Output Operator [RS_56] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_49] + | outputColumnNames:["_col0"] + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_115] + | predicate:((d_year = 1998) and d_date_sk is not null) (type: boolean) + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_47] + | alias:date_dim + | Statistics:Num rows: 73049 Data size: 81741831 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 6 [SIMPLE_EDGE] + Reduce Output Operator [RS_54] + key expressions:_col0 (type: int) + Map-reduce partition columns:_col0 (type: int) + sort order:+ + Statistics:Num rows: 16500000 Data size: 16745518417 Basic stats: COMPLETE Column stats: NONE + value expressions:_col4 (type: int), _col11 (type: int), _col13 (type: decimal(7,2)), _col14 (type: decimal(7,2)) + Select Operator [SEL_46] + outputColumnNames:["_col0","_col11","_col13","_col14","_col4"] + Statistics:Num rows: 16500000 Data size: 16745518417 Basic stats: COMPLETE Column stats: NONE + Filter Operator [FIL_107] + predicate:(((_col23) IN ('KY', 'GA', 'NM') and _col6 BETWEEN 100 AND 200) or ((_col23) IN ('MT', 'OR', 'IN') and _col6 BETWEEN 150 AND 300) or ((_col23) IN ('WI', 'MO', 'WV') and _col6 BETWEEN 50 AND 250)) (type: boolean) + Statistics:Num rows: 16500000 Data size: 16745518417 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_121] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col9 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col4","_col6","_col11","_col13","_col14","_col23"] + | Statistics:Num rows: 22000000 Data size: 22327357890 Basic stats: COMPLETE Column stats: NONE + |<-Map 15 [SIMPLE_EDGE] + | Reduce Output Operator [RS_43] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 20000000 Data size: 20297597642 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: string) + | Select Operator [SEL_34] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 20000000 Data size: 20297597642 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_114] + | predicate:((((ca_state) IN ('KY', 'GA', 'NM') or (ca_state) IN ('MT', 'OR', 'IN') or (ca_state) IN ('WI', 'MO', 'WV')) and (ca_country = 'United States')) and ca_address_sk is not null) (type: boolean) + | Statistics:Num rows: 20000000 Data size: 20297597642 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_32] + | alias:customer_address + | Statistics:Num rows: 40000000 Data size: 40595195284 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 5 [SIMPLE_EDGE] + Reduce Output Operator [RS_41] + key expressions:_col9 (type: int) + Map-reduce partition columns:_col9 (type: int) + sort order:+ + Statistics:Num rows: 21780 Data size: 7888165 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: int), _col4 (type: int), _col6 (type: decimal(7,2)), _col11 (type: int), _col13 (type: decimal(7,2)), _col14 (type: decimal(7,2)) + Merge Join Operator [MERGEJOIN_120] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col10 (type: int), _col17 (type: string), _col18 (type: string)","1":"_col0 (type: int), _col1 (type: string), _col2 (type: string)"} + | outputColumnNames:["_col0","_col4","_col6","_col9","_col11","_col13","_col14"] + | Statistics:Num rows: 21780 Data size: 7888165 Basic stats: COMPLETE Column stats: NONE + |<-Map 14 [SIMPLE_EDGE] + | Reduce Output Operator [RS_38] + | key expressions:_col0 (type: int), _col1 (type: string), _col2 (type: string) + | Map-reduce partition columns:_col0 (type: int), _col1 (type: string), _col2 (type: string) + | sort order:+++ + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_31] + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_113] + | predicate:((((((cd_education_status = '4 yr Degree') or (cd_education_status = 'Primary') or (cd_education_status = 'Advanced Degree')) and cd_education_status is not null) and ((cd_marital_status = 'M') or (cd_marital_status = 'D') or (cd_marital_status = 'U'))) and cd_marital_status is not null) and cd_demo_sk is not null) (type: boolean) + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_29] + | alias:cd1 + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 4 [SIMPLE_EDGE] + Reduce Output Operator [RS_36] + key expressions:_col10 (type: int), _col17 (type: string), _col18 (type: string) + Map-reduce partition columns:_col10 (type: int), _col17 (type: string), _col18 (type: string) + sort order:+++ + Statistics:Num rows: 8166 Data size: 2957518 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: int), _col4 (type: int), _col6 (type: decimal(7,2)), _col9 (type: int), _col11 (type: int), _col13 (type: decimal(7,2)), _col14 (type: decimal(7,2)) + Select Operator [SEL_28] + outputColumnNames:["_col0","_col10","_col11","_col13","_col14","_col17","_col18","_col4","_col6","_col9"] + Statistics:Num rows: 8166 Data size: 2957518 Basic stats: COMPLETE Column stats: NONE + Filter Operator [FIL_108] + predicate:(((_col17 = 'M') and (_col18 = '4 yr Degree') and _col5 BETWEEN 100.0 AND 150.0) or ((_col17 = 'D') and (_col18 = 'Primary') and _col5 BETWEEN 50.0 AND 100.0) or ((_col17 = 'U') and (_col18 = 'Advanced Degree') and _col5 BETWEEN 150.0 AND 200.0)) (type: boolean) + Statistics:Num rows: 8166 Data size: 2957518 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_119] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col8 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col4","_col5","_col6","_col9","_col10","_col11","_col13","_col14","_col17","_col18"] + | Statistics:Num rows: 21780 Data size: 7888165 Basic stats: COMPLETE Column stats: NONE + |<-Map 13 [SIMPLE_EDGE] + | Reduce Output Operator [RS_25] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: string), _col2 (type: string) + | Select Operator [SEL_11] + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_112] + | predicate:((((((cd_marital_status = 'M') or (cd_marital_status = 'D') or (cd_marital_status = 'U')) and ((cd_education_status = '4 yr Degree') or (cd_education_status = 'Primary') or (cd_education_status = 'Advanced Degree'))) and cd_demo_sk is not null) and cd_education_status is not null) and cd_marital_status is not null) (type: boolean) + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_9] + | alias:cd1 + | Statistics:Num rows: 19800 Data size: 7171059 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 3 [SIMPLE_EDGE] + Reduce Output Operator [RS_23] + key expressions:_col8 (type: int) + Map-reduce partition columns:_col8 (type: int) + sort order:+ + Statistics:Num rows: 5062 Data size: 2965795 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: int), _col4 (type: int), _col5 (type: decimal(7,2)), _col6 (type: decimal(7,2)), _col9 (type: int), _col10 (type: int), _col11 (type: int), _col13 (type: decimal(7,2)), _col14 (type: decimal(7,2)) + Merge Join Operator [MERGEJOIN_118] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col2 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col0","_col4","_col5","_col6","_col8","_col9","_col10","_col11","_col13","_col14"] + | Statistics:Num rows: 5062 Data size: 2965795 Basic stats: COMPLETE Column stats: NONE + |<-Map 12 [SIMPLE_EDGE] + | Reduce Output Operator [RS_20] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 4602 Data size: 2696178 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_8] + | outputColumnNames:["_col0"] + | Statistics:Num rows: 4602 Data size: 2696178 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_111] + | predicate:wp_web_page_sk is not null (type: boolean) + | Statistics:Num rows: 4602 Data size: 2696178 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_6] + | alias:web_page + | Statistics:Num rows: 4602 Data size: 2696178 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 2 [SIMPLE_EDGE] + Reduce Output Operator [RS_18] + key expressions:_col2 (type: int) + Map-reduce partition columns:_col2 (type: int) + sort order:+ + Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + value expressions:_col0 (type: int), _col4 (type: int), _col5 (type: decimal(7,2)), _col6 (type: decimal(7,2)), _col8 (type: int), _col9 (type: int), _col10 (type: int), _col11 (type: int), _col13 (type: decimal(7,2)), _col14 (type: decimal(7,2)) + Merge Join Operator [MERGEJOIN_117] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col1 (type: int), _col3 (type: int)","1":"_col0 (type: int), _col5 (type: int)"} + | outputColumnNames:["_col0","_col2","_col4","_col5","_col6","_col8","_col9","_col10","_col11","_col13","_col14"] + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + |<-Map 1 [SIMPLE_EDGE] + | Reduce Output Operator [RS_13] + | key expressions:_col1 (type: int), _col3 (type: int) + | Map-reduce partition columns:_col1 (type: int), _col3 (type: int) + | sort order:++ + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | value expressions:_col0 (type: int), _col2 (type: int), _col4 (type: int), _col5 (type: decimal(7,2)), _col6 (type: decimal(7,2)) + | Select Operator [SEL_2] + | outputColumnNames:["_col0","_col1","_col2","_col3","_col4","_col5","_col6"] + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | Filter Operator [FIL_109] + | predicate:(((((ws_item_sk is not null and (ws_sales_price BETWEEN 100.0 AND 150.0 or ws_sales_price BETWEEN 50.0 AND 100.0 or ws_sales_price BETWEEN 150.0 AND 200.0)) and ws_order_number is not null) and (ws_net_profit BETWEEN 100 AND 200 or ws_net_profit BETWEEN 150 AND 300 or ws_net_profit BETWEEN 50 AND 250)) and ws_web_page_sk is not null) and ws_sold_date_sk is not null) (type: boolean) + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | TableScan [TS_0] + | alias:web_sales + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + |<-Map 11 [SIMPLE_EDGE] + Reduce Output Operator [RS_15] + key expressions:_col0 (type: int), _col5 (type: int) + Map-reduce partition columns:_col0 (type: int), _col5 (type: int) + sort order:++ + Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + value expressions:_col1 (type: int), _col2 (type: int), _col3 (type: int), _col4 (type: int), _col6 (type: decimal(7,2)), _col7 (type: decimal(7,2)) + Select Operator [SEL_5] + outputColumnNames:["_col0","_col1","_col2","_col3","_col4","_col5","_col6","_col7"] + Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + Filter Operator [FIL_110] + predicate:(((((wr_order_number is not null and wr_item_sk is not null) and wr_refunded_cdemo_sk is not null) and wr_returning_cdemo_sk is not null) and wr_refunded_addr_sk is not null) and wr_reason_sk is not null) (type: boolean) + Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + TableScan [TS_3] + alias:web_returns + Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + http://git-wip-us.apache.org/repos/asf/hive/blob/09b6f9a3/ql/src/test/results/clientpositive/perf/query87.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/perf/query87.q.out b/ql/src/test/results/clientpositive/perf/query87.q.out new file mode 100644 index 0000000..e4762b2 --- /dev/null +++ b/ql/src/test/results/clientpositive/perf/query87.q.out @@ -0,0 +1,334 @@ +PREHOOK: query: explain select count(*) from (select distinct c_last_name as l1, c_first_name as f1, d_date as d1 from store_sales JOIN date_dim ON store_sales.ss_sold_date_sk = date_dim.d_date_sk JOIN customer ON store_sales.ss_customer_sk = customer.c_customer_sk where d_month_seq between 1193 and 1193+11 ) t1 LEFT OUTER JOIN ( select distinct c_last_name as l2, c_first_name as f2, d_date as d2 from catalog_sales JOIN date_dim ON catalog_sales.cs_sold_date_sk = date_dim.d_date_sk JOIN customer ON catalog_sales.cs_bill_customer_sk = customer.c_customer_sk where d_month_seq between 1193 and 1193+11 ) t2 ON t1.l1 = t2.l2 and t1.f1 = t2.f2 and t1.d1 = t2.d2 LEFT OUTER JOIN (select distinct c_last_name as l3, c_first_name as f3, d_date as d3 from web_sales JOIN date_dim ON web_sales.ws_sold_date_sk = date_dim.d_date_sk JOIN customer ON web_sales.ws_bill_customer_sk = customer.c_customer_sk where d_month_seq between 1193 and 1193+11 ) t3 ON t1.l1 = t3.l3 and t1.f1 = t3.f3 and t1.d1 = t3 .d3 WHERE l2 is null and l3 is null +PREHOOK: type: QUERY +POSTHOOK: query: explain select count(*) from (select distinct c_last_name as l1, c_first_name as f1, d_date as d1 from store_sales JOIN date_dim ON store_sales.ss_sold_date_sk = date_dim.d_date_sk JOIN customer ON store_sales.ss_customer_sk = customer.c_customer_sk where d_month_seq between 1193 and 1193+11 ) t1 LEFT OUTER JOIN ( select distinct c_last_name as l2, c_first_name as f2, d_date as d2 from catalog_sales JOIN date_dim ON catalog_sales.cs_sold_date_sk = date_dim.d_date_sk JOIN customer ON catalog_sales.cs_bill_customer_sk = customer.c_customer_sk where d_month_seq between 1193 and 1193+11 ) t2 ON t1.l1 = t2.l2 and t1.f1 = t2.f2 and t1.d1 = t2.d2 LEFT OUTER JOIN (select distinct c_last_name as l3, c_first_name as f3, d_date as d3 from web_sales JOIN date_dim ON web_sales.ws_sold_date_sk = date_dim.d_date_sk JOIN customer ON web_sales.ws_bill_customer_sk = customer.c_customer_sk where d_month_seq between 1193 and 1193+11 ) t3 ON t1.l1 = t3.l3 and t1.f1 = t3.f3 and t1.d1 = t 3.d3 WHERE l2 is null and l3 is null +POSTHOOK: type: QUERY +Plan optimized by CBO. + +Vertex dependency in root stage +Reducer 11 <- Map 10 (SIMPLE_EDGE), Map 14 (SIMPLE_EDGE) +Reducer 12 <- Map 15 (SIMPLE_EDGE), Reducer 11 (SIMPLE_EDGE) +Reducer 13 <- Reducer 12 (SIMPLE_EDGE) +Reducer 17 <- Map 16 (SIMPLE_EDGE), Map 20 (SIMPLE_EDGE) +Reducer 18 <- Map 21 (SIMPLE_EDGE), Reducer 17 (SIMPLE_EDGE) +Reducer 19 <- Reducer 18 (SIMPLE_EDGE) +Reducer 2 <- Map 1 (SIMPLE_EDGE), Map 8 (SIMPLE_EDGE) +Reducer 3 <- Map 9 (SIMPLE_EDGE), Reducer 2 (SIMPLE_EDGE) +Reducer 4 <- Reducer 3 (SIMPLE_EDGE) +Reducer 5 <- Reducer 13 (SIMPLE_EDGE), Reducer 4 (SIMPLE_EDGE) +Reducer 6 <- Reducer 19 (SIMPLE_EDGE), Reducer 5 (SIMPLE_EDGE) +Reducer 7 <- Reducer 6 (SIMPLE_EDGE) + +Stage-0 + Fetch Operator + limit:-1 + Stage-1 + Reducer 7 + File Output Operator [FS_86] + compressed:false + Statistics:Num rows: 1 Data size: 8 Basic stats: COMPLETE Column stats: NONE + table:{"input format:":"org.apache.hadoop.mapred.TextInputFormat","output format:":"org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat","serde:":"org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe"} + Group By Operator [GBY_84] + | aggregations:["count(VALUE._col0)"] + | outputColumnNames:["_col0"] + | Statistics:Num rows: 1 Data size: 8 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 6 [SIMPLE_EDGE] + Reduce Output Operator [RS_83] + sort order: + Statistics:Num rows: 1 Data size: 8 Basic stats: COMPLETE Column stats: NONE + value expressions:_col0 (type: bigint) + Group By Operator [GBY_82] + aggregations:["count()"] + outputColumnNames:["_col0"] + Statistics:Num rows: 1 Data size: 8 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_81] + Statistics:Num rows: 24200000 Data size: 20812489029 Basic stats: COMPLETE Column stats: NONE + Filter Operator [FIL_108] + predicate:_col6 is null (type: boolean) + Statistics:Num rows: 24200000 Data size: 20812489029 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_126] + | condition map:[{"":"Left Outer Join0 to 1"}] + | keys:{"0":"_col0 (type: string), _col1 (type: string), _col2 (type: string)","1":"_col0 (type: string), _col1 (type: string), _col2 (type: string)"} + | outputColumnNames:["_col6"] + | Statistics:Num rows: 48400001 Data size: 41624978920 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 19 [SIMPLE_EDGE] + | Reduce Output Operator [RS_78] + | key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | sort order:+++ + | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_76] + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + | Group By Operator [GBY_75] + | | keys:KEY._col0 (type: string), KEY._col1 (type: string), KEY._col2 (type: string) + | | outputColumnNames:["_col0","_col1","_col2"] + | | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + | |<-Reducer 18 [SIMPLE_EDGE] + | Reduce Output Operator [RS_74] + | key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | sort order:+++ + | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + | Group By Operator [GBY_73] + | keys:_col3 (type: string), _col6 (type: string), _col7 (type: string) + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + | Merge Join Operator [MERGEJOIN_124] + | | condition map:[{"":"Inner Join 0 to 1"}] + | | keys:{"0":"_col1 (type: int)","1":"_col0 (type: int)"} + | | outputColumnNames:["_col3","_col6","_col7"] + | | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + | |<-Map 21 [SIMPLE_EDGE] + | | Reduce Output Operator [RS_70] + | | key expressions:_col0 (type: int) + | | Map-reduce partition columns:_col0 (type: int) + | | sort order:+ + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | | value expressions:_col1 (type: string), _col2 (type: string) + | | Select Operator [SEL_61] + | | outputColumnNames:["_col0","_col1","_col2"] + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | | Filter Operator [FIL_118] + | | predicate:c_customer_sk is not null (type: boolean) + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | | TableScan [TS_59] + | | alias:customer + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | |<-Reducer 17 [SIMPLE_EDGE] + | Reduce Output Operator [RS_68] + | key expressions:_col1 (type: int) + | Map-reduce partition columns:_col1 (type: int) + | sort order:+ + | Statistics:Num rows: 40176 Data size: 44957392 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col3 (type: string) + | Merge Join Operator [MERGEJOIN_123] + | | condition map:[{"":"Inner Join 0 to 1"}] + | | keys:{"0":"_col0 (type: int)","1":"_col0 (type: int)"} + | | outputColumnNames:["_col1","_col3"] + | | Statistics:Num rows: 40176 Data size: 44957392 Basic stats: COMPLETE Column stats: NONE + | |<-Map 16 [SIMPLE_EDGE] + | | Reduce Output Operator [RS_63] + | | key expressions:_col0 (type: int) + | | Map-reduce partition columns:_col0 (type: int) + | | sort order:+ + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | | value expressions:_col1 (type: int) + | | Select Operator [SEL_55] + | | outputColumnNames:["_col0","_col1"] + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | | Filter Operator [FIL_116] + | | predicate:(ws_sold_date_sk is not null and ws_bill_customer_sk is not null) (type: boolean) + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | | TableScan [TS_53] + | | alias:web_sales + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | |<-Map 20 [SIMPLE_EDGE] + | Reduce Output Operator [RS_65] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: string) + | Select Operator [SEL_58] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_117] + | predicate:(d_month_seq BETWEEN 1193 AND 1204 and d_date_sk is not null) (type: boolean) + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_56] + | alias:date_dim + | Statistics:Num rows: 73049 Data size: 81741831 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 5 [SIMPLE_EDGE] + Reduce Output Operator [RS_77] + key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + sort order:+++ + Statistics:Num rows: 24200000 Data size: 20812489029 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_52] + outputColumnNames:["_col0","_col1","_col2"] + Statistics:Num rows: 24200000 Data size: 20812489029 Basic stats: COMPLETE Column stats: NONE + Filter Operator [FIL_109] + predicate:_col3 is null (type: boolean) + Statistics:Num rows: 24200000 Data size: 20812489029 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_125] + | condition map:[{"":"Left Outer Join0 to 1"}] + | keys:{"0":"_col0 (type: string), _col1 (type: string), _col2 (type: string)","1":"_col0 (type: string), _col1 (type: string), _col2 (type: string)"} + | outputColumnNames:["_col0","_col1","_col2","_col3"] + | Statistics:Num rows: 48400001 Data size: 41624978920 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 13 [SIMPLE_EDGE] + | Reduce Output Operator [RS_49] + | key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | sort order:+++ + | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + | Select Operator [SEL_47] + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + | Group By Operator [GBY_46] + | | keys:KEY._col0 (type: string), KEY._col1 (type: string), KEY._col2 (type: string) + | | outputColumnNames:["_col0","_col1","_col2"] + | | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + | |<-Reducer 12 [SIMPLE_EDGE] + | Reduce Output Operator [RS_45] + | key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + | sort order:+++ + | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + | Group By Operator [GBY_44] + | keys:_col3 (type: string), _col6 (type: string), _col7 (type: string) + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + | Merge Join Operator [MERGEJOIN_122] + | | condition map:[{"":"Inner Join 0 to 1"}] + | | keys:{"0":"_col1 (type: int)","1":"_col0 (type: int)"} + | | outputColumnNames:["_col3","_col6","_col7"] + | | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + | |<-Map 15 [SIMPLE_EDGE] + | | Reduce Output Operator [RS_41] + | | key expressions:_col0 (type: int) + | | Map-reduce partition columns:_col0 (type: int) + | | sort order:+ + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | | value expressions:_col1 (type: string), _col2 (type: string) + | | Select Operator [SEL_32] + | | outputColumnNames:["_col0","_col1","_col2"] + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | | Filter Operator [FIL_115] + | | predicate:c_customer_sk is not null (type: boolean) + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | | TableScan [TS_30] + | | alias:customer + | | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | |<-Reducer 11 [SIMPLE_EDGE] + | Reduce Output Operator [RS_39] + | key expressions:_col1 (type: int) + | Map-reduce partition columns:_col1 (type: int) + | sort order:+ + | Statistics:Num rows: 40176 Data size: 44957392 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col3 (type: string) + | Merge Join Operator [MERGEJOIN_121] + | | condition map:[{"":"Inner Join 0 to 1"}] + | | keys:{"0":"_col0 (type: int)","1":"_col0 (type: int)"} + | | outputColumnNames:["_col1","_col3"] + | | Statistics:Num rows: 40176 Data size: 44957392 Basic stats: COMPLETE Column stats: NONE + | |<-Map 10 [SIMPLE_EDGE] + | | Reduce Output Operator [RS_34] + | | key expressions:_col0 (type: int) + | | Map-reduce partition columns:_col0 (type: int) + | | sort order:+ + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | | value expressions:_col1 (type: int) + | | Select Operator [SEL_26] + | | outputColumnNames:["_col0","_col1"] + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | | Filter Operator [FIL_113] + | | predicate:(cs_sold_date_sk is not null and cs_bill_customer_sk is not null) (type: boolean) + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | | TableScan [TS_24] + | | alias:catalog_sales + | | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | |<-Map 14 [SIMPLE_EDGE] + | Reduce Output Operator [RS_36] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: string) + | Select Operator [SEL_29] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_114] + | predicate:(d_month_seq BETWEEN 1193 AND 1204 and d_date_sk is not null) (type: boolean) + | Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_27] + | alias:date_dim + | Statistics:Num rows: 73049 Data size: 81741831 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 4 [SIMPLE_EDGE] + Reduce Output Operator [RS_48] + key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + sort order:+++ + Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + Select Operator [SEL_23] + outputColumnNames:["_col0","_col1","_col2"] + Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + Group By Operator [GBY_22] + | keys:KEY._col0 (type: string), KEY._col1 (type: string), KEY._col2 (type: string) + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 44000000 Data size: 37840889108 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 3 [SIMPLE_EDGE] + Reduce Output Operator [RS_21] + key expressions:_col0 (type: string), _col1 (type: string), _col2 (type: string) + Map-reduce partition columns:_col0 (type: string), _col1 (type: string), _col2 (type: string) + sort order:+++ + Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + Group By Operator [GBY_20] + keys:_col3 (type: string), _col6 (type: string), _col7 (type: string) + outputColumnNames:["_col0","_col1","_col2"] + Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + Merge Join Operator [MERGEJOIN_120] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col1 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col3","_col6","_col7"] + | Statistics:Num rows: 88000001 Data size: 75681779077 Basic stats: COMPLETE Column stats: NONE + |<-Map 9 [SIMPLE_EDGE] + | Reduce Output Operator [RS_17] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | value expressions:_col1 (type: string), _col2 (type: string) + | Select Operator [SEL_8] + | outputColumnNames:["_col0","_col1","_col2"] + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | Filter Operator [FIL_112] + | predicate:c_customer_sk is not null (type: boolean) + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + | TableScan [TS_6] + | alias:customer + | Statistics:Num rows: 80000000 Data size: 68801615852 Basic stats: COMPLETE Column stats: NONE + |<-Reducer 2 [SIMPLE_EDGE] + Reduce Output Operator [RS_15] + key expressions:_col1 (type: int) + Map-reduce partition columns:_col1 (type: int) + sort order:+ + Statistics:Num rows: 40176 Data size: 44957392 Basic stats: COMPLETE Column stats: NONE + value expressions:_col3 (type: string) + Merge Join Operator [MERGEJOIN_119] + | condition map:[{"":"Inner Join 0 to 1"}] + | keys:{"0":"_col0 (type: int)","1":"_col0 (type: int)"} + | outputColumnNames:["_col1","_col3"] + | Statistics:Num rows: 40176 Data size: 44957392 Basic stats: COMPLETE Column stats: NONE + |<-Map 1 [SIMPLE_EDGE] + | Reduce Output Operator [RS_10] + | key expressions:_col0 (type: int) + | Map-reduce partition columns:_col0 (type: int) + | sort order:+ + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | value expressions:_col1 (type: int) + | Select Operator [SEL_2] + | outputColumnNames:["_col0","_col1"] + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | Filter Operator [FIL_110] + | predicate:(ss_sold_date_sk is not null and ss_customer_sk is not null) (type: boolean) + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + | TableScan [TS_0] + | alias:store_sales + | Statistics:Num rows: 1 Data size: 0 Basic stats: PARTIAL Column stats: NONE + |<-Map 8 [SIMPLE_EDGE] + Reduce Output Operator [RS_12] + key expressions:_col0 (type: int) + Map-reduce partition columns:_col0 (type: int) + sort order:+ + Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + value expressions:_col1 (type: string) + Select Operator [SEL_5] + outputColumnNames:["_col0","_col1"] + Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + Filter Operator [FIL_111] + predicate:(d_month_seq BETWEEN 1193 AND 1204 and d_date_sk is not null) (type: boolean) + Statistics:Num rows: 36524 Data size: 40870356 Basic stats: COMPLETE Column stats: NONE + TableScan [TS_3] + alias:date_dim + Statistics:Num rows: 73049 Data size: 81741831 Basic stats: COMPLETE Column stats: NONE +
