Question

I have the following (fairly legacy, did this a year or two ago) SQL query. The webpage the SQL is on (used on an SqlDataSource/GridView in ASP.NET) is very slow, and I’ve pinpointed the slowness to this query – seemingly because of the sub select. I’ve tried using joins to hopefully speed it up but I can’t get it to work. Any ideas?

I won’t put the whole query here, simply because I work on a machine with no internet access so I won’t be able to copy and paste, and most of it is just selecting from the main table.

SELECT ..., 
       CASE 
         WHEN di.Total = di.Delivered THEN 'Received' 
         ELSE 'Not Received' END AS 'Status',
       ...
FROM Deliveries AS d 
  LEFT OUTER JOIN (
         SELECT Delivery, 
                COUNT(*) AS Total, 
                COUNT(CASE WHEN Status = 2 THEN 1 END) AS Delivered 
         FROM DeliveryItems 
         GROUP BY Delivery
  ) AS di ON d.ID = di.Delivery

Any tips?

Answer 1

SELECT Delivery, 
       COUNT(*) AS Total, 
       COUNT(CASE WHEN Status = 2 THEN 1 END) AS Delivered 
FROM DeliveryItems 
GROUP BY Delivery

This part could be improved by ditching that CASE statement. Maybe try joining twice, once against the total count and once against the filtered count.

SELECT ..., 
       CASE 
         WHEN d2.Total = d1.Delivered THEN 'Received' 
         ELSE 'Not Received' END AS 'Status',
       ...
FROM Deliveries AS d 

LEFT OUTER JOIN (

SELECT Delivery,
       COUNT(*) AS Delivered
FROM DeliveryItems
WHERE Status = 2
GROUP BY Delivery ) d1 ON d.Id = d1.Delivery

LEFT OUTER JOIN (

SELECT Delivery,
       COUNT(*) AS Total
FROM DeliveryItems
GROUP BY Delivery ) d2 ON d.ID = d2.Delivery