Question

I have a table with roughly 7,000,000 records.

It’s very flat and for sake of argument it has 3 columns I wish to aggregate on. This aggregation should very simply create a count /pivot of each instance of that value.

E.G

Company     Status  Year
Hatstand    Open    2011
Hatstand    Closed  2011
Moonbase    Open    2011

Would produce

Count of Hatstand **2**
Count of Hatstand Open **1**
Count of Hatstand Open 2011 **1**

So it’s a very simple count of each “branch” of data.

My first choice was to use a SSRS Matrix control. Which when testing with a small dataset worked really well. However when using a the full data-set would not run.

What is the “correct” way to approach this problem?

Should I pre-aggregate via a stored procedure or SSIS job?
Or should I continue with the SSRS route and try to refine my query?

Thanks

Answer 1

T-SQL is preferred way to go if You are using SQL Server 2005 or 2008.

If You are using SQL Server 2005 or 2008 You can try:

SELECT *, COUNT(*) FROM TBL
GROUP BY COMPANY, STATUS, YEAR
WITH ROLLUP -- or WITH CUBE

If You are using SQL Server 2008 You can try with :

SELECT *, COUNT(*) FROM A
GROUP BY 
GROUPING SETS (
    (COMPANY),
    (COMPANY, STATUS),
    (COMPANY, STATUS, YEAR)
)

For details about GROUP BY WITH ROLLUP/CUBE and GROUPING SETS take a look at GROUP BY.