How to query a large table using update command with multiple joins? - sql

Question

I have a big table on an external server, and I am trying do an update on another table on a local machine. The external server table is more than a couple of billion rows. When I run the following update as a stored procedure, it takes a very long time to give results. Is there a method to tweak this so that it can be broken into smaller datasets to query?
update C set C.[P_Flag]='Y' from [LocalMachine].dbo.O_C_Data C inner
join [ExternalServer].dbo.E_Extract E on C.[O_ID]=E.[B_ID] and
C.[P_Option]=E.[P_OPTID] and C.[P]=E.[PD_G] where C.[O_Flag]='Y'

Answer 1

EDIT Just as #usr pointed out, what I said below is not correct. You'd better know why it is slow by checking execution plan, which I didn't master when I faced the same problem. You query might trigger too many executions to remote server. That is to query data on remote server separately for every single row on local. However, I suggest to have a try with the first approach below. It did solve similar problem before. I suspect that you have the same root cause.
When you join local table to a table in remote server, SQL Server data engine will pull all data from table in remote and do JOIN in your local memory. That's why it's extremely slow.
I suggest creating a local temp table, inserting temp table with data pulled from remote server and updating local table with temp table. To achieve that, you still need to JOIN local table and remote table. There are 2 approaches:
Embed local table in query. In your case, can you write
INSERT INTO #TempTable
SELECT E.B_ID
FROM [ExternalServer].dbo.E_Extract E
WHERE E.[B_ID] IN (C_ID1, C_ID2, ...) AND E.[P_OPTID] IN (P_Option1, P_Option2) AND E.[PD_G] IN (P1, P2, ...)
Note, values after IN operator are constants.
Use REMOTE JOIN Hints. Take a look at MSDN
REMOTE
Specifies that the join operation is performed on the site of the right table. This is useful when the left table is a local table and the right table is a remote table. REMOTE should be used only when the left table has fewer rows than the right table.
If the right table is local, the join is performed locally. If both tables are remote but from different data sources, REMOTE causes the join to be performed on the site of the right table. If both tables are remote tables from the same data source, REMOTE is not required.
REMOTE cannot be used when one of the values being compared in the join predicate is cast to a different collation using the COLLATE clause.
REMOTE can be used only for INNER JOIN operations.