胖子哥的大数据之路（10）- 基于Hive构建数据仓库实例

一、引言

　　基于Hive+Hadoop模式构建数据仓库，是大数据时代的一个不错的选择，本文以郑商所每日交易行情数据为案例，探讨数据Hive数据导入的操作实例。

二、源数据-每日行情数据

三、建表脚本

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

CREATE TABLE IF NOT EXISTS t_day_detail( id STRING, lastday FLOAT, today FLOAT, highest FLOAT, lowest FLOAT, today_end FLOAT, today_jisuan FLOAT, updown1 FLOAT, updown2 FLOAT, sum int, empity int, rise int, turnover FLOAT, delivery FLOAT ) PARTITIONED BY (dt STRING,product STRING);

 四、数据导入1

load data local inpath '/home/hadoop/source/in'
overwrite into table t_day_detail
partition(dt='2014-04-22',product='1');

五、数据导入2

load data local inpath '/home/hadoop/source/in'
overwrite into table t_day_detail
partition(dt='2014-04-23',product='1');

六、执行结果

hive> select * from t_day_detail
    > ;
OK
CF405,17365.0,17390.0,17390.0,17360.0,17380.0,17380.0,15,15,72.0,1090.0,-36,625.66,0.0  NULL    NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-22      1
CF407,17275.0,17370.0,17415.0,17320.0,17320.0,17365.0,45,90,22.0,52.0,2,191.01,0.0      NULL    NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-22      1
CF409,17450.0,17380.0,17395.0,17310.0,17320.0,17330.0,-130,-120,7860.0,34584.0,-940,68099.08,0.0        NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-22      1
CF411,16370.0,16315.0,16350.0,16220.0,16255.0,16240.0,-115,-130,984.0,17436.0,-380,7990.01,0.0  NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-22      1
CF501,16130.0,16030.0,16085.0,15920.0,15995.0,15970.0,-135,-160,26210.0,115120.0,-1906,209311.56,0.0    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-22      1
CF503,16195.0,16030.0,16065.0,16000.0,16065.0,16045.0,-130,-150,60.0,526.0,12,481.42,0.0        NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-22      1
CF405,17365.0,17390.0,17390.0,17360.0,17380.0,17380.0,15,15,72.0,1090.0,-36,625.66,0.0  NULL    NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-23      1
CF407,17275.0,17370.0,17415.0,17320.0,17320.0,17365.0,45,90,22.0,52.0,2,191.01,0.0      NULL    NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-23      1
CF409,17450.0,17380.0,17395.0,17310.0,17320.0,17330.0,-130,-120,7860.0,34584.0,-940,68099.08,0.0        NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-23      1
CF411,16370.0,16315.0,16350.0,16220.0,16255.0,16240.0,-115,-130,984.0,17436.0,-380,7990.01,0.0  NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-23      1
CF501,16130.0,16030.0,16085.0,15920.0,15995.0,15970.0,-135,-160,26210.0,115120.0,-1906,209311.56,0.0    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-23      1
CF503,16195.0,16030.0,16065.0,16000.0,16065.0,16045.0,-130,-150,60.0,526.0,12,481.42,0.0        NULL    NULL    NULL    NULLNULL     NULL    NULL    NULL    NULL    NULL    NULL    NULL    NULL    2014-04-23      1
Time taken: 0.391 seconds
hive> 

七、问题聚焦

（1）Hive分区设置的原则和技巧？

（2）Hive分区内行记录完全一致问题如何解决？

（3）Hive中文乱码问题如何解决？