declare @i int
set @i = 0
while @i<500000
begin
	print(getdate())
	if @i%10000=0 
	begin
		print(@i)
	end
	set @i = @i + 1
end

insert into table_1 (col1,col2,col3,col4)
select 1,2,3,4  union all
select 12,2,3,4 union all
select 132,3,4  union all
select 14,2,3,4 union all
select 15,2,3,4 union all
...etc...

quote:

---

Originally posted by Zathras

The only way to access data from a certain design database is to write PML code that send the data to my SQL Server. The PML function basically opens a connection, send 300,000+ INSERT statements one-by-one, then executes a SP to tell my SQL Server that it has finished exporting. This SP then moves the data from the one set of tables to the "real" tables so that end users can see the complete data. This process of exporting takes about 8 to 9 hours per project becuase it is done in PML code (no choice).

Sometimes that export stalls due to say the exporting computer crashes. To check for this, the SQL Server runs a SP every 20 minutes looking for projects that hasn't written data in over 20 minutes and doesn't have the finish marker. They way I check for data is that every single INSERT has a TimeStamp DateTime field as the very last field and the data is inserted as a GETDATE().

My question is, is there any other way I can get the last time data was written for a certain project (connection?) other than having a separate TimeStamp DateTime field. I'm trying to speed up the exports since they take so long. Not having 300,000+ GETDATE() function calls seems to be a good place to start.

EDIT: The field is a datetime field (not the TimeStamp datatype)

---