Question

我想生成一些随机数据。 ~50k行

每行都有ID自动数字和10到1的数字值。

+----+------+------+------+------+------+------+------+------+------+------+
| id | nr1  | nr2  | nr3  | nr4  | nr5  | nr6  | nr7  | nr8  | nr9  | nr10 |
+----+------+------+------+------+------+------+------+------+------+------+
| 1  | 1    | 39   | 19   | 23   | 28   | 80   | 3    | 42   | 60   | 32   |
+----+------+------+------+------+------+------+------+------+------+------+
| 2  | 43   | 18   | 3    | 24   | 29   | 33   | 15   | 1    | 61   | 80   |
+----+------+------+------+------+------+------+------+------+------+------+
| 3  | 11   | 25   | 33   | 2    | 30   | 3    | 1    | 44   | 62   | 78   |
+----+------+------+------+------+------+------+------+------+------+------+

因此每行需要有10个不同的值。

我可以生成10个随机数，但比较它们之间的所有10个，以确保没有重复值需要花费很多时间。

INSERT INTO test
SELECT trunc(random() * 89 + 1) as nr1,
       trunc(random() * 89 + 1) as nr2,
       trunc(random() * 89 + 1) as nr3,
       trunc(random() * 89 + 1) as nr4,
       trunc(random() * 89 + 1) as nr5,
       trunc(random() * 89 + 1) as nr6,
       trunc(random() * 89 + 1) as nr7,
       trunc(random() * 89 + 1) as nr8,
       trunc(random() * 89 + 1) as nr9,
       trunc(random() * 89 + 1) as nr10
FROM generate_series(1,50);

那么执行该任务的更有效方法是什么。

Answer 1

为什么不随机化您想要的范围内的数字顺序（比如1,100），然后使用前10个。

循环和枢轴将执行此操作：

declare @result table (id int, nr1 int, nr2 int, nr3 int, nr4 int, nr5 int, nr6 int, nr7 int, nr8 int, nr9 int, nr10 int)

declare @numberSet table (n int)
declare @i table (rw int,rnk int, n int)

insert into @numberset
SELECT DISTINCT n = number 
FROM master..[spt_values] 
WHERE number BETWEEN 1 AND 100


declare @l int
set @l = 0

WHILE (@l < 50000)
BEGIN
    insert into @i
    select top 10 @l+1,rank() over (order by id), n from (select newid() id , n from @numberset ) b
    SELECT @l = @l + 1
END

INSERT INTO @result
SELECT  rw,[1],[2],[3],[4],[5],[6],[7],[8],[9],[10]
FROM (SELECT rw, n,rnk from @i) as sourcetable
PIVOT
(max(n) for rnk in ([1],[2],[3],[4],[5],[6],[7],[8],[9],[10])
) as pivottable;

SELECT * FROM @result

Answer 2

尝试：

-- INSERT INTO table( ...... )
CREATE TABLE test111 AS
SELECT 
       max( case when rn = 1 then nr end ) nr1,
       max( case when rn = 2 then nr end ) nr2,
       max( case when rn = 3 then nr end ) nr3,
       max( case when rn = 4 then nr end ) nr4,
       max( case when rn = 5 then nr end ) nr5,
       max( case when rn = 6 then nr end ) nr6,
       max( case when rn = 7 then nr end ) nr7,
       max( case when rn = 8 then nr end ) nr8,
       max( case when rn = 9 then nr end ) nr9,
       max( case when rn = 10 then nr end ) nr10
FROM generate_series(1,50000) id,
LATERAL ( 
  SELECT nr,
         row_number() over ( ORDER BY id * random() ) rn
  FROM generate_series(1,90) nr
)  x
GROUP BY id

此CREATE TABLE查询在我的电脑上播放大约8-9秒：
Query OK, 50000 rows affected (execution time: 8,016 sec; total time: 8,016 sec)

编辑 - PostgreSQL版本的版本＆lt; = 9.2

CREATE TABLE test222 AS
SELECT        
       max( case when rn = 1 then nr end ) nr1,
       max( case when rn = 2 then nr end ) nr2,
       max( case when rn = 3 then nr end ) nr3,
       max( case when rn = 4 then nr end ) nr4,
       max( case when rn = 5 then nr end ) nr5,
       max( case when rn = 6 then nr end ) nr6,
       max( case when rn = 7 then nr end ) nr7,
       max( case when rn = 8 then nr end ) nr8,
       max( case when rn = 9 then nr end ) nr9,
       max( case when rn = 10 then nr end ) nr10
FROM (
    SELECT id, nr, row_number() over (partition by id order by random()) rn 
    FROM generate_series(1,50000) id
    CROSS JOIN (
      SELECT *  FROM generate_series(1,90) nr
    ) x
) y
WHERE rn <= 10
GROUP BY id

这个版本在我的电脑上大约需要7-8秒：
Query OK, 50000 rows affected (execution time: 7,516 sec; total time: 7,516 sec)

创建10个不同的随机数

2 个答案: