事故指标统计

[[email protected] ~]# hive

Logging initialized using configuration in jar:file:/usr/local/soft/hive-1.2.1/lib/hive-common-1.2.1.jar!/hive-log4j.properties
hive> select sgfssj from dwd.base_acd_file order by sgfssj limit 10;
2006-01-22 09:30:00.0
2006-12-21 00:15:00.0
2006-12-21 13:50:00.0
2006-12-21 16:30:00.0
2006-12-21 18:02:00.0
2006-12-22 11:30:00.0
2006-12-22 13:30:00.0
2006-12-22 14:30:00.0
2006-12-22 17:30:00.0
2006-12-22 23:55:00.0
Time taken: 28.559 seconds, Fetched: 10 row(s)
hive> select sgfssj from dwd.base_acd_file order by sgfssj desc limit 10;
OK
2020-10-14 17:28:00.0
2020-10-14 11:16:00.0
2020-10-13 23:06:00.0
2020-10-13 19:25:00.0
2020-10-13 14:12:00.0
2020-10-13 12:10:00.0
2020-10-12 16:04:00.0
2020-10-12 11:50:00.0
2020-10-12 07:38:00.0
2020-10-12 07:30:00.0
Time taken: 21.673 seconds, Fetched: 10 row(s)
hive> select current_date;
OK
2022-07-05
Time taken: 0.459 seconds, Fetched: 1 row(s)
hive> select year(sgfssj),count(*) from dwd.base_acd_file group by year(sgfssj);
2006	43
2007	1082
2008	1070
2009	1377
2010	1579
2011	2604
2012	2117
2013	1802
2014	1936
2015	1991
2016	2094
2017	1933
2018	2373
2019	2617
2020	1930
Time taken: 22.075 seconds, Fetched: 15 row(s)
hive> select substr(sgfssj,1,10)
    >         ,count(1)as dr_sgs
    > from dwd.base_acd_file 
    > group by substr(sgfssj,1,10);
Time taken: 45.99 seconds, Fetched: 4966 row(s)
hive> select t1.tjrq
    >         ,t1.dr_sgs
    >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
    > from(
    >     select substr(sgfssj,1,10)as tjrq
    >         ,count(1)as dr_sgs
    >     from dwd.base_acd_file 
    >     group by substr(sgfssj,1,10)
    > )t1;
Time taken: 45.99 seconds, Fetched: 4966 row(s)

hive> select t1.tjrq
    >         ,t1.dr_sgs
    >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
    >         ,lag(t1.dr_sgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_sgs
    > from(
    >     select substr(sgfssj,1,10)as tjrq
    >         ,count(1)as dr_sgs
    >     from dwd.base_acd_file 
    >     group by substr(sgfssj,1,10)
    > )t1;
Time taken: 73.903 seconds, Fetched: 4966 row(s)
hive> exit;
[[email protected] ~]# cd /usr/local/soft/spark-2.4.5/
[[email protected] spark-2.4.5]# ls
bin   examples    LICENSE   NOTICE  README.md  yarn
conf  jars        licenses  python  RELEASE
data  kubernetes  logs      R       sbin
[[email protected] spark-2.4.5]# spark-sql --conf spark.sql.shuffle.partitions=2
spark-sql> show databases;
22/07/05 11:11:24 INFO codegen.CodeGenerator: Code generated in 227.508135 ms
default
dwd
Time taken: 0.351 seconds, Fetched 2 row(s)
22/07/05 11:11:24 INFO thriftserver.SparkSQLCLIDriver: Time taken: 0.351 seconds, Fetched 2 row(s)
spark-sql> use dwd;
Time taken: 0.041 seconds
22/07/05 11:11:31 INFO thriftserver.SparkSQLCLIDriver: Time taken: 0.041 seconds
spark-sql> show tables;
22/07/05 11:11:41 INFO spark.ContextCleaner: Cleaned accumulator 2
22/07/05 11:11:41 INFO spark.ContextCleaner: Cleaned accumulator 0
22/07/05 11:11:41 INFO spark.ContextCleaner: Cleaned accumulator 1
22/07/05 11:11:41 INFO codegen.CodeGenerator: Code generated in 13.604985 ms
dwd	base_acd_file	false
dwd	base_acd_filehuman	false
dwd	base_bd_drivinglicense	false
dwd	base_bd_vehicle	false
dwd	base_vio_force	false
dwd	base_vio_surveil	false
dwd	base_vio_violation	false
Time taken: 0.119 seconds, Fetched 7 row(s)
22/07/05 11:11:41 INFO thriftserver.SparkSQLCLIDriver: Time taken: 0.119 seconds, Fetched 7 row(s)

spark-sql> select tt1.tjrq
         >         ,tt1.dr_sgs
         >         ,tt1.jn_sgs
         >         ,tt1.qntq_sgs
         >         ,NVL(round((abs(tt1.dr_sgs-tt1.qntq_sgs)/NVL(tt1.qntq_sgs,1))*100,2),0) as tb_sgs
         >         ,if(tt1.dr_sgs-tt1.qntq_sgs>0,"上升",'下降') as tb_sgs_bj
         > from(
         >     select t1.tjrq
         >         ,t1.dr_sgs
         >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
         >         ,lag(t1.dr_sgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_sgs
         >     from(
         >         select substr(sgfssj,1,10)as tjrq
         >             ,count(1)as dr_sgs
         >         from dwd.base_acd_file 
         >         group by substr(sgfssj,1,10)
         > )t1
         > )tt1;
Time taken: 0.596 seconds, Fetched 4966 row(s)
22/07/05 11:32:08 INFO thriftserver.SparkSQLCLIDriver: Time taken: 0.596 seconds, Fetched 4966 row(s)
spark-sql> select tt1.tjrq
         >         ,tt1.dr_sgs
         >         ,tt1.jn_sgs
         >         ,tt1.qntq_sgs
         >         ,NVL(round((abs(tt1.dr_sgs-tt1.qntq_sgs)/NVL(tt1.qntq_sgs,1))*100,2),0) as tb_sgs
         >         ,if(tt1.dr_sgs-tt1.qntq_sgs>0,"上升",'下降') as tb_sgs_bj
         > from(
         >     select t1.tjrq
         >         ,t1.dr_sgs
         >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
         >         ,lag(t1.dr_sgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_sgs
         >     from(
         >         select substr(sgfssj,1,10)as tjrq
         >             ,count(1)as dr_sgs
         >         from dwd.base_acd_file 
         >         group by substr(sgfssj,1,10)
         > )t1
         > )tt1
         > order by tt1.tjrq;
Time taken: 5.46 seconds, Fetched 4966 row(s)
22/07/05 14:36:35 INFO thriftserver.SparkSQLCLIDriver: Time taken: 5.46 seconds, Fetched 4966 row(s)

spark-sql> select tt1.tjrq
         >         ,tt1.dr_sgs
         >         ,tt1.jn_sgs
         >         ,tt1.qntq_sgs
         > 
         >         ,NVL(round((abs(tt1.dr_sgs-tt1.qntq_sgs)/NVL(tt1.qntq_sgs,1))*100,2),0) as tb_sgs
         >         ,if(tt1.dr_sgs-tt1.qntq_sgs>0,"上升",'下降') as tb_sgs_bj
         > 
         >         ,NVL(round((abs(tt1.dr_swsgs-tt1.qntq_swsgs)/NVL(tt1.qntq_swsgs,1))*100,2),0) as tb_swsgs
         >         ,if(tt1.dr_swsgs-tt1.qntq_swsgs>0,"上升",'下降') as tb_swsgs_bj
         > from(
         >     select t1.tjrq
         >         ,t1.dr_sgs
         >         ,t1.dr_swsgs
         > 
         >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
         >         ,lag(t1.dr_sgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_sgs
         > 
         >         ,sum(t1.dr_swsgs)over (partition by substr(t1.tjrq,1,4))as jn_swsgs
         >         ,lag(t1.dr_swsgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_swsgs
         >     from(
         >         select substr(sgfssj,1,10)as tjrq
         >             ,count(1)as dr_sgs
         >             ,sum(if(swrs7>0,1,0))as dr_swsgs
         >         from dwd.base_acd_file 
         >         group by substr(sgfssj,1,10)
         > )t1
         > )tt1
         > order by tt1.tjrq;
Time taken: 5.372 seconds, Fetched 4966 row(s)
22/07/05 15:17:19 INFO thriftserver.SparkSQLCLIDriver: Time taken: 5.372 seconds, Fetched 4966 row(s)

spark-sql> select tt1.tjrq
         >         ,tt1.dr_sgs
         >         ,tt1.jn_sgs
         >         ,tt1.qntq_sgs
         >         ,tt1.dr_swsgs
         >         ,tt1.qntq_swsgs
         > 
         >         ,NVL(round((abs(tt1.dr_sgs-tt1.qntq_sgs)/NVL(tt1.qntq_sgs,1))*100,2),0) as tb_sgs
         >         ,if(tt1.dr_sgs-tt1.qntq_sgs>0,"上升",'下降') as tb_sgs_bj
         > 
         >         ,NVL(round((abs(tt1.dr_swsgs-tt1.qntq_swsgs)/NVL(tt1.qntq_swsgs,1))*100,2),0) as tb_swsgs
         >         ,if(tt1.dr_swsgs-tt1.qntq_swsgs>0,"上升",'下降') as tb_swsgs_bj
         > from(
         >     select t1.tjrq
         >         ,t1.dr_sgs
         >         ,t1.dr_swsgs
         > 
         >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
         >         ,lag(t1.dr_sgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_sgs
         > 
         >         ,sum(t1.dr_swsgs)over (partition by substr(t1.tjrq,1,4))as jn_swsgs
         >         ,lag(t1.dr_swsgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_swsgs
         >     from(
         >         select substr(sgfssj,1,10)as tjrq
         >             ,count(1)as dr_sgs
         >             ,sum(if(swrs7>0,1,0))as dr_swsgs
         >         from dwd.base_acd_file 
         >         group by substr(sgfssj,1,10)
         > )t1
         > )tt1
         > order by tt1.tjrq;
Time taken: 4.13 seconds, Fetched 4966 row(s)
22/07/05 15:20:00 INFO thriftserver.SparkSQLCLIDriver: Time taken: 4.13 seconds, Fetched 4966 row(s)

spark-sql> select tt1.tjrq
         >         ,tt1.dr_sgs
         >         ,tt1.jn_sgs
         >         ,tt1.qntq_sgs
         >         ,tt1.dr_swsgs
         >         ,tt1.jn_swsgs
         >         ,tt1.qntq_swsgs
         > 
         >         ,NVL(round((abs(tt1.dr_sgs-tt1.qntq_sgs)/NVL(tt1.qntq_sgs,1))*100,2),0) as tb_sgs
         >         ,if(tt1.dr_sgs-tt1.qntq_sgs>0,"上升",'下降') as tb_sgs_bj
         > 
         >         ,NVL(round((abs(tt1.dr_swsgs-tt1.qntq_swsgs)/NVL(tt1.qntq_swsgs,1))*100,2),0) as tb_swsgs
         >         ,if(tt1.dr_swsgs-tt1.qntq_swsgs>0,"上升",'下降') as tb_swsgs_bj
         > from(
         >     select t1.tjrq
         >         ,t1.dr_sgs
         >         ,t1.dr_swsgs
         > 
         >         ,sum(t1.dr_sgs)over (partition by substr(t1.tjrq,1,4))as jn_sgs
         >         ,lag(t1.dr_sgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_sgs
         > 
         >         ,sum(t1.dr_swsgs)over (partition by substr(t1.tjrq,1,4))as jn_swsgs
         >         ,lag(t1.dr_swsgs,1,1)over (partition by substr(t1.tjrq,6,5)order by substr(t1.tjrq,1,4))as qntq_swsgs
         >     from(
         >         select substr(sgfssj,1,10)as tjrq
         >             ,count(1)as dr_sgs
         >             ,sum(if(swrs7>0,1,0))as dr_swsgs
         >         from dwd.base_acd_file 
         >         group by substr(sgfssj,1,10)
         > )t1
         > )tt1
         > order by tt1.tjrq;
Time taken: 5.653 seconds, Fetched 4966 row(s)
22/07/05 15:23:09 INFO thriftserver.SparkSQLCLIDriver: Time taken: 5.653 seconds, Fetched 4966 row(s)