检查数据是否已经存在并插入的更好方法

Question

Hi all I have the following SQL Server 2008 script that will check if a row already exists in Table A and if it doesn't insert the data from Table B .

This was working nicely until Table A started to fill up with a lot of data. We currently have 30 million rows in this table and this will continue to grow to a predicated 70 million rows.

The problem if this is taking far too long and is affecting other processes. Just wondering if there is a better way to check if a row already exists in a table. Just to add as well this is all done using an SSIS.

Script:

INSERT INTO TABLE A ([recordID],Field 1, Field2, Field 3, Field 4, Field 5) 
    SELECT 
        [TABLE B].[recordID],[TABLE B].[Field 1], [TABLE B].[Field2], 
        [TABLE B].[Field 3], [TABLE B].[Field 4], [TABLE B].[Field 5] 
    FROM TABLE B AS TABLE B 
    LEFT OUTER JOIN TABLE A AS TABLE A ON [TABLE B].[recordID] = [TABLE A].[recordID] 
    WHERE [TABLE A].[recordID] IS NULL

Answer 1

You can check the the Merge command:

http://technet.microsoft.com/en-us/library/bb510625.aspx

Answer 2

Not sure if this will be faster, but worth a try:

INSERT INTO TABLE A ([recordID],Field 1, Field2, Field 3, Field 4, Field 5)

SELECT [TABLE B].[recordID],[TABLE B].[Field 1], [TABLE B].[Field2], [TABLE B].[Field 3], [TABLE B].[Field 4], [TABLE B].[Field 5] 
FROM TABLE B 
where b.recordID not in 
     (select recordID from A)

Answer 3

1. If it is clustered Index, and the newly added RecordId is not incremental, then lot of page spilt is expected to happen. Make sure to set optimal Fill Factor.
2. Find what operation takes time, then it will be simple to address. If it Searching is taking time or inserting is taking time?