Does an INSTEAD OF trigger gets re-created every time you insert into a view?

Question

I have SQL Server 2016. I have two tables with the same structure, one contains historical data, and the other one contains current data.
I have a view that programs use, that returns the content of both tables.

CREATE TABLE A (int Id IDENTITY(1,1) NOT NULL, othercolumns, primary key, ect.)
CREATE TABLE B (int Id NOT NULL, othercolumns, primary key, ect.)
CREATE VIEW MyView AS SELECT * FROM A UNION ALL SELECT * FROM B

CREATE TRIGGER TR_INSTEADOFINSERT_ON_MyView on MyView
INSTEAD OF INSERT
AS
BEGIN
INSERT INTO A (columns)
   SELECT data
   FROM inserted
END 

I was wondering if the view gets re-created everytime an insert is done in the view. Because if I check in the sys.dm_exec_sessions and sys.db_exec_requests, I always see the "CREATE TRIGGER TR_INSTEADOFINSERT" in the query text beign executed.

Now I know that it's taking a long time because it's a big table and I have a lot of activity on the server. But I was just wondering why I keep seeing the "CREATE TRIGGER" in the sql_text of the query being executed? Is it re-created or not? What's the most efficient way to do this? Should I modify the programs to reference the table directly when inserting?

Answer 1

The trigger is not being recreated. sql_text in sys.db_exec_requests is at the batch level, not the query level. See eg

https://blogs.msdn.microsoft.com/dbrowne/2013/01/18/my-favorite-query-for-investigating-sql-server-performance/

For an example of how to parse out the currently-running query from the larger batch.

That wait is to fetch the page that the row is written to, either on the Clustered Index or some non-clustered index. So I would look at the index design of the table, and see if it should be better-optimized for inserts.