Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
Author |
Topic |
naveendkt
Starting Member
5 Posts |
Posted - 2011-08-19 : 16:24:39
|
I have a Query something like below SELECT S.StoreID,S.LocationName,ISNULL(tmp7.WithoutChild,'0'),ISNULL(tmp3.ThreeToEight,'0')FROM Store S LEFT JOIN (SELECT COUNT(DISTINCT R.ReservationID) AS WithoutChild,R.StoreID FROM Reservation R INNER JOIN Customer C ON C.CustomerID = R.CustomerID WHERE (R.ReservationID NOT IN(SELECT ReservationID FROM ReservationChild)) AND (R.StartingTime BETWEEN '08/15/2011' AND '08/19/2011')GROUP BY R.StoreID) AS tmp7 ON tmp7.StoreID = S.StoreIDLEFT JOIN (SELECT COUNT(DISTINCT R.ReservationID) AS ThreeToEight,R.StoreID FROM Reservation R INNER JOIN Customer C ON C.CustomerID = R.CustomerID INNER JOIN ReservationChild RC ON RC.ReservationID = R.ReservationID WHERE (R.StartingTime BETWEEN '08/15/2011' AND '08/19/2011') AND (RC.Age BETWEEN 9 AND 12)GROUP BY R.StoreID) AS tmp3 ON tmp3.StoreID = S.StoreIDI have added only 2 left joins here,Actual query has 7. Each is a temporary table with COUNT() as one of its column.This query times out when the time range is given large. BTW this query id running inside a SP. Do I have any option of decreasing the execution time for this SP?Other Info: Reservation table has around 1819320 rows and Store table has 210. FYI:The query will return rows in Store table and columns are Count with different conditions from Reservation table. Naveen |
|
visakh16
Very Important crosS Applying yaK Herder
52326 Posts |
Posted - 2011-08-20 : 01:54:23
|
try to merge it like this and seeSELECT S.StoreID,S.LocationName,ISNULL(tmp7.WithoutChild,'0'),ISNULL(tmp7.ThreeToEight,'0')FROM Store S LEFT JOIN (SELECT COUNT(DISTINCT R.ReservationID) AS WithoutChild,COUNT(DISTINCT CASE WHEN RC.ReservationID IS NOT NULLAND RC.Age BETWEEN 9 AND 12 THEN RC.ReservationID ELSE NULL END) AS ThreeToEight,R.StoreID FROM Reservation R INNER JOIN Customer C ON C.CustomerID = R.CustomerIDLEFT JOIN ReservationChild RC ON RC.ReservationID = R.ReservationID WHERE (R.StartingTime BETWEEN '08/15/2011' AND '08/19/2011')GROUP BY R.StoreID) AS tmp7 ON tmp7.StoreID = S.StoreID also why do you need DISTINCT inside count? will reservationid repeat for customer? also if you dont want to return any customer info why include it in join then?------------------------------------------------------------------------------------------------------SQL Server MVPhttp://visakhm.blogspot.com/ |
 |
|
visakh16
Very Important crosS Applying yaK Herder
52326 Posts |
Posted - 2011-08-20 : 01:55:08
|
Also check whether you've required indexes on joining columns by analysing execution plan.------------------------------------------------------------------------------------------------------SQL Server MVPhttp://visakhm.blogspot.com/ |
 |
|
naveendkt
Starting Member
5 Posts |
Posted - 2011-08-20 : 13:46:43
|
Thanks visakh16 for your reply. For your first question. No the reservationid won't repeat in Reservation table. I have removed the Distinct inside COUNT since its a primary key.I have Inner joined Customer because there are reservations without customerid, i don't want those reservations. There is index defined for all the columns which is used to join. I am going to try as you suggested soon. And I will let you know the result. Also as I told I am using this inside SP. So if i use temporary tables to get the counts one by one will that help? Since there is left joins used to get each count in my query, is that affecting the execution time?Thank you,NaveenNaveen |
 |
|
visakh16
Very Important crosS Applying yaK Herder
52326 Posts |
Posted - 2011-08-20 : 13:57:42
|
no need of even temporary table. you can first do main insert and then use a series of updates to get other count values one by one------------------------------------------------------------------------------------------------------SQL Server MVPhttp://visakhm.blogspot.com/ |
 |
|
|
|
|
|
|