基于HDFS

没有专门的数据存储格式

存储结构主要包括：数据库、文件、表、视图

可以直接加载文本文件（.txt文件）

创建表时，指定Hive数据的列分隔符与行分隔符

Hive的数据模型

表

Table 内部表

hive> create table t1
    > (tid int,tname string,age int);
hive> create table t2
    > (tid int,tname string,age int)
    > location '/mytable/hive/t2';
hive> create table t3
    > (tid int,tname string,age int)
    > row format delimited fields terminated by ',';
hive>create table t3
    >(tid int,tname string,age int)
    >row format delimited fields terminated by ',';
hive>create table t4
    >as
    >select * from wordcount;
hive>select * from t4;
#hdfs dfs -ls /user/hive/warehouse/t4/
#hdfs dfs -cat /user/hive/warehouse/t4/000000-0
hive>create table t5
    >row format delimited fields terminated by ','
    >as
    >select * from wordcount;
#hdfs dfs -cat /user/hive/warehouse/t5/000000-0
hive>alter table t1 add columns(english int);
#desc t1;

与数据库中的 Table 在概念上是类似

每一个 Table 在 Hive 中都有一个相应的目录存储数据

所有的 Table 数据（不包括 External Table）都保存在这个目录中

删除表时，元数据与数据都会被删除

Partition 分区表

Partition 对应于数据库的 Partition 列的密集索引

在 Hive 中，表中的一个 Partition 对应于表下的一个目录，所有的 Partition 的数据都存储在对应的目录中

hive>create table partition_table
    >(sid int,sname string)
    >partitioned by (gender string)
    >row format delimited fields terminated by ',';
hive>insert into table partition_table partition(gender='M') select sid,sname from wordcountwhere gender='M';
hive>insert into table partition_table partition(gender='F') select sid,sname from wordcountwhere gender='F';

hive>explain select * from sample_data where gender='M';
hive>explain select * from partition_table where gender='M';

External Table 外部表

指向已经在 HDFS 中存在的数据，可以创建 Partition

它和内部表在元数据的组织上是相同的，而实际数据的存储则有较大的差异

外部表 只有一个过程，加载数据和创建表同时完成，并不会移动到数据仓库目录中，只是与外部数据建立一个链接。当删除一个外部表时，仅删除该链接

#vi student01.txt
1,Tom,23
2,Mary,20
#vi student02.txt
3,Mike,25
#vi student03.txt
4,Scott,21
5,King,20
#hdfs dfs -put student01.txt /input
#hdfs dfs -put student02.txt /input
#hdfs dfs -put student03.txt /input
hive> create external table external_student
    > (sid int,sname string,age int)
    > row format delimited fields terminated by ','
    > location '/input';

hive>select * from external_student;
# hdfs dfs -rm /input/student03.txt
hive>select * from external_student;
#hdfs dfs -put student03.txt /input
hive>select * from external_student;

Bucket Table 桶表

桶表是对数据进行哈希取值，然后放到不同文件中存储。

hive> create table bucket_table
    > (sid int,sname string,age int)
    > clustered by(sname) into 5 buckets;

视图

视图是一种虚表，是一个逻辑概念；可以跨越多张表

视图建立在已有表的基础上, 视图赖以建立的这些表称为基表

视图可以简化复杂的查询

方法一
hive>create table dept (
       >deptno int,
       >dname string)
       >row format delimited fields terminated by ' ';
hive>insert into table dept
       >values
       >(10.0,'SALES'),
       >(20.0,'RESEARCH'),
       >(30.0,'ACCOUTING');
方法二
#vi a.txt
10  SALES
20  RESEARCH
30  ACCOUTING
# hdfs dfs -mkdir /dd
# hdfs dfs -put emp.txt /dd

hvie>create table dept
       >(deptno int,dname string)
       >row format delimited fields terminated by ' ';

hive>load data inpath '/dd' overwrite into table dept;

hive>select * from dept;

hive>create table emp
       >(empno int,
       >ename string,
       >job string,
       >mgr string,
       >hiredate date,
       >sal int,
       >comm int,
       >deptno int)
       >row format delimited fields terminated by ' ';

hive>insert into table emp
       >values
       >(7369.0,'SMITH','CLERK',7902,'1980-12-17',800,'NULL',20),
       >(7499.0,'ALLEN','SALESMAN',7698,'1981-02-20',1600,300,30),
       >(7521.0,'WARD','SALESMAN',7698,'1981-02-22',1250,500,30),
       >(7566.0,'JONES','MANAGER',7839,'1981-04-02',2975,'NULL',20),
       >(7654.0,'MARTIN','SALESMAN',7698,'1981-09-28',1250,1400,30),
       >(7698.0,'BLAKE','MANAGER',7839,'1981-05-01',2850,'NULL',30),
       >(7782.0,'CLARK','MANAGER',7839,'1981-06-09',2450,'NULL',10),
       >(7788.0,'SCOTT','ANALYST',7566,'1987-04-19',3000,'NULL',20);

hive>insert into table emp
       >(empno,ename,job,hiredate,sal,comm,deptno)
       >values
       >(7839.0,'KING','PRESIDENT','1981-11-17',5000,'NULL',10);
hive>insert into table emp
       >values
       >(7844.0,'TURNER','SALESMAN',7698,'1981-09-08',1500,0,30),
       >(7876.0,'ADAMS','CLERK',7788,'1987-05-23',1100,'NULL',20),
       >(7900.0,'JAMES','CLERK',7698,'1981-12-03',950,'NULL',30),
       >(7902.0,'FORD','ANALYST',7566,'1981-12-03',3000,'NULL',20),
       >(7934.0,'MILLER','CLERK',7782,'1982-01-23',1300,'NULL',10);

--查询员工信息：员工号，姓名，月薪，年薪，部门名称
hive>create view empinfo
       >as 
       >select e.empno,e.ename,e.sal,e.sal*12 annlsal,d.dname
       >from emp e,dept d
       >where e.deptno=d.deptno;
hive>desc empinfo;
hive>select * from empinfo;