如何使查询结果分区为输入?

时间:2019-04-04 22:47:17

标签: hive hiveql

我是蜂巢的新手,所以有一个基本问题:如何创建查询以使查询结果以特定方式划分?

例如:

CREATE TABLE IF NOT EXISTS tbl_x (
 x SMALLINT,
 y FLOAT)
PARTITIONED BY (id SMALLINT)
ROW FORMAT DELIMITED
FIELDS TERMINATED BY ','
STORED AS ORC;

INSERT INTO TABLE `tbl_x` 
VALUES (1, 1, 1.0),
       (1, 1, 2.0),
       (1, 2, 3.0),
       (1, 2, 4.0),
       (2, 1, 5.0),
       (2, 1, 6.0),
       (2, 2, 7.0),
       (2, 2, 8.0);

CREATE TABLE tbl_y AS SELECT `id`, `x`, SUM(`y`) AS `y_sum`
FROM `tbl_x`
GROUP BY `id`, `x`;

在该示例中,我也希望对tbl_y进行分区。

尝试不起作用

CREATE TABLE tbl_y AS SELECT `id`, `x`, SUM(`y`) AS `y_sum`
FROM `tbl_x`
GROUP BY `id`, `x` PARTITIONED BY (id SMALLINT);  

这里的窍门是什么?我应该先定义分区表并将结果插入其中吗?

1 个答案:

答案 0 :(得分:1)

是的,您应该创建一个分区表,因为不支持select(CTAS)创建分区表。

CREATE TABLE tbl_y(x smallint,y_sum double)
partitioned by (id smallint)
STORED AS ORC;

如果表架构相同,则可以使用CREATE LIKE

CREATE TABLE tbl_y like tbl_x;

您还可以使用DISTRIBUTE BY在减速器之间平均分配数据,另请参见以下答案:https://stackoverflow.com/a/38475807/2700344

insert overwrite table tbl_y partition(id)
select id, x, SUM(y) AS y_sum
fromtbl_x
group by id, x 
distribute by id, FLOOR(RAND()*100.0)%20;