如何在 Azure Synapse 中为标识列设置种子值和增量值？

Question

我在 Azure Synapse 数据仓库中有一个 table 和 DDL，如下所示：

CREATE TABLE [trans_customer_cdm_ejkb].[cdm_file_process_history]
(
    [id] [int] IDENTITY(1,1) NOT NULL,
    [layer] [varchar](500) NOT NULL,
    [ingest_partition] [varchar](100) NULL,
    [status] [varchar](25) NULL,
    [last_update_time] [datetime2](7) NULL,
    [pipeline_run_id] [varchar](200) NULL
)
WITH
(
    DISTRIBUTION = HASH ( [ingest_partition] ),
    CLUSTERED COLUMNSTORE INDEX
)
GO

我尝试使用下面的 sql 将值附加到此 table 3 次，id 列的种子值为 45，增量值为 60，尽管 IDENTITY(1,1 ).

DECLARE @now DATETIME2 = GETDATE()

INSERT INTO trans_customer_cdm_ejkb.cdm_file_process_history
(layer, ingest_partition, [status], last_update_time, pipeline_run_id)
VALUES ('src2stg', '2022-02-18-03', 'success', @now, 'Test')

SELECT * FROM trans_customer_cdm_ejkb.cdm_file_process_history

此外，我使用下面的 SQL 检查种子值为 1，增量值为 1。但是，table 没有按预期提供 id 值

SELECT  sm.name
,       tb.name
,       co.name
,       ic.seed_value
,       ic.increment_value
FROM        sys.schemas AS sm
JOIN        sys.tables  AS tb           ON  sm.schema_id = tb.schema_id
JOIN        sys.columns AS co           ON  tb.object_id = co.object_id
JOIN        sys.identity_columns AS ic  ON  co.object_id = ic.object_id
                                        AND co.column_id = ic.column_id
WHERE   sm.name = 'trans_customer_cdm_ejkb'
AND     tb.name = 'cdm_file_process_history'
;

我该如何解决这个问题？

亲切的问候，

Answer 1

IDENTITY Azure Synapse Analytics 专用 SQL 池中的列确实保证唯一值，但 do not guarantee sequential values。原因是因为数据分布在 60 个分布中；每个分布都有一组唯一的标识值。

如果有一个连续的列对你很重要，那么重新创建 table 而没有 IDENTITY 属性并将你的 INSERT 语句更改为以下代码，这将生成顺序 ID：


DECLARE @now DATETIME2 = GETDATE()

INSERT INTO trans_customer_cdm_ejkb.cdm_file_process_history
(id, layer, ingest_partition, [status], last_update_time, pipeline_run_id)
SELECT
(SELECT ISNULL(MAX(id),0) FROM trans_customer_cdm_ejkb.cdm_file_process_history) +1 as id, 
 'src2stg', '2022-02-18-03', 'success', @now, 'Test'

SELECT * FROM trans_customer_cdm_ejkb.cdm_file_process_history

因为你的代码只是插入一行，所以我做了 +1 但通常你会做 + ROW_NUMBER() OVER (ORDER BY [SomeColumn]).

如何在 Azure Synapse 中为标识列设置种子值和增量值？

How to set seed value and incremental value for identity column in Azure Synapse?

sql

data-warehouse

azure-synapse