Not quite sure what you mean by 'parameter 1' but:

declare @bigvarchar varchar(1000)

set @bigvarchar = @name_of_variable

J

try putting square brackets around the object names:

select * from [ST_Rel_4.050_Test].[dbo].[sa_desk]

You probably only need it around the database name but I have shown it around all levels.

J

A bookmark lookup is when the query uses a non-clustered index to finds rows matching and then uses a clustered index to find the row in the table.

If you are...

I would load the data into a single table and then write queries from that table to insert into the normalised tables.

I'm not sure that you can do it all...

Have a look at http://www.w3schools.com/ado/

set conn=Server.CreateObject("ADODB.Connection")

conn.Provider="Microsoft.Jet.OLEDB.4.0"

conn.Open "c:/webdata/northwind.mdb"

set rs=Server.CreateObject("ADODB.recordset")rs.Open "exec stored_proce 'start_date', 'end_date'", conn

There are other ways of doing it like using command type and adding parameters but I cannot...

It is doing a full table scan because there is no usable index.  AccountId is part of one index but it is not the first column in the index and...

Same as you would for return rows from a table:

create procedure Myproc @start_date datetime, @end_date datetime

select <columns>

from vw_summary

where sdate between @start_date and @end_date

J

As far as the error message, there is no way round it.  Every column in the select has to be either in the Group By or part of an aggregate...

Have you got some data/ddl you can share with us?

J

Where do the running totals come in?

J

Might be the cause of the problem!

Any other way of getting the SentTo information?

J

Check the size of the files first either using xp_cmdshell 'dir \\server\folder\*.txt'.  You can put these results into a table and query that to determine the size.  You could...

I guess that it is a 1 to Many between fn_getValidSubscriptions and tblcommunicationsSent. 

The query is doing a distinct on tblCommunicationsSent.SentTo - and the number of Many in the relationship...

With the ZIP data, why cant you load the character data into an int column - wont SQL Server will do an implicit cast during the load?

J

Would vertical partitioning be of any benefit as you cannot normalise the data?  If columns SN10 - SN60 are empty for most rows then you might not need to have...