This will copy all distinct rows from one table, create a new table and insert the distinct rows only.

how can we improve query performence. with same query dont use indexes view . for example select * from mytable query execute 2 min but the same query i want execute with in 1 min. hw can we do that one , please give me reply

SELECT * into DupRecordsTbl_bkp FROM
(SELECT * FROM DupRecordsTbl INTERSECT SELECT * FROM DupRecordsTbl)B
DELETE FROM DupRecordsTbl
INSERT INTO DupRecordsTbl SELECT * FROM DupRecordsTbl_bkp

Keeps the unique record into a new table. Deletes the original table records and reinsert the unique records into original table.

Here we need no info about columns as it compares the whole row.

delete from table where table.%%physloc%% not in (select MIN(%%physloc%%) from table
group by duplicate columns)

CustomerID Name Address …
2101 ABC 123 blaa
2101 WQH 123 blaa

The Name and Address should remain the same but the CustomerID must be changed

Thank you

I am in a situation where I have the same CustomerID’s but different Customer names. They belong to the same Subsidiary Entity so the phone numbers and address must remain the same..

Please help

Below is the table with sample data that I have

Date Name Product Quantity
10/11/2011 Aname Abc1 20
10/10/2011 Aname Abc1 20
10/9/2011 Aname Abc1 20
10/8/2011 AName Xyz1 30
10/7/2011 AName Xyz1 20
10/6/2011 AName Abc1 20
10/5/2011 AName Abc1 20
10/4/2011 BName Xyz1 30
10/11/2011 Bname Abc2 20
10/10/2011 Bname Abc2 20
10/9/2011 Bname Abc2 20
10/8/2011 BName Xyz2 30
10/7/2011 BName Xyz2 20
10/6/2011 BName Abc2 20
10/5/2011 BName Abc2 20
10/4/2011 BName Xyz2 30

Output should be :

Id Start Date EndDate Name Product Quantity
1 10/9/2011 10/11/2011 Aname Abc1 20
2 10/8/2011 10/9/2011 Aname Xyz1 30
3 10/7/2011 10/8/2011 Aname Xyz1 20
4 10/5/2011 10/7/2011 Aname Abc1 20
5 10/4/2011 10/5/2011 Aname Xyz1 30
6 10/9/2011 10/11/2011 Bname Abc2 20
7 10/8/2011 10/9/2011 Bname Xyz2 30
8 10/7/2011 10/8/2011 Bname Xyz2 20
9 10/5/2011 10/7/2011 Bname Abc2 20
10 10/4/2011 10/5/2011 Bname Xyz2 30

I do not have primary key on my Table.

If you see the number of rows have been reduced.

I have more than 500 million records and by filtering this I would significantly the number of records in the table.

Is there any efficient query or sproc to do this, if so could you please help me.

I have tried using group by , partitions, ranking but nothing worked .

Thanks & Regards,

VPandit.

Riketa Salariya & Dhiresh jawale

Simply you rock..

Select Distinct Row_Number() Over (Order By CandidateM.CanID Asc)As SlNo, CanCode,CanName from CandidateM Where CandidateM.CanCode like ’09%’
Using this query how can I take rownumber of a distinct record….?
regards,

josekutty varghese

INSERT INTO BankMaster
VALUES
(1,’ING Vysya’,'ING Vysya Eranamkulam’),
(2,’SBT’,'SBT Eranamkulam’),
(3,’Federal’,'Federal Bank Eranamkulam’),
(4,’Canara’,'Canara Eranamkulam’),
(5,’ING Vysya’,'ING Vysya Eranamkulam’),
(6,’SIB’,'SIB Eranamkulam’),
(7,’SBT’,'SBT Eranamkulam’),
(8,’Canara’,'Canara Eranamkulam’),
(9,’ING Vysya’,'ING Vysya Eranamkulam’),
(10,’SBI’,'SBI Eranamkulam’)

DECLARE @BankName NVARCHAR(50),@Cnt INT

DECLARE Cursor_Bank CURSOR LOCAL FAST_FORWARD READ_ONLY FOR
SELECT BankName,COUNT(BankName) – 1 from BankMaster
GROUP BY BankName HAVING COUNT(BankName) > 1
OPEN Cursor_Bank
FETCH NEXT FROM Cursor_Bank INTO @bankname,@Cnt
WHILE (@@FETCH_STATUS = 0)
BEGIN
SET ROWCOUNT @Cnt
DELETE FROM BankMaster WHERE BankName like @bankname
SET ROWCOUNT 0
FETCH NEXT FROM Cursor_Bank INTO @bankname,@Cnt
END
CLOSE Cursor_Bank
DEALLOCATE Cursor_Bank

If i am wrong then correct me.
As per identity column in a table u will never have a duplicate rows in a table
because it grows as 1,2,3,4,5
so,we can’nt have duplicate rows in a identity column.
but,we can have duplicate data in other column of a identity tables.

Find another method using loop.

Consider the table #temp as below,
Id Value
——————
1 100
2 200
2 200
3 300
3 300
3 300

DECLARE @Cnt INT
SET @Cnt = 1

WHILE @Cnt IS NOT NULL
BEGIN
SELECT TOP 1 @Cnt = COUNT(Id) FROM #Test
GROUP BY Id,Value
HAVING COUNT(Id) > 1
ORDER BY Id,Value

IF @Cnt IS NULL OR @Cnt = 1 RETURN

SET @Cnt = @Cnt-1
SET ROWCOUNT @Cnt

DELETE t1
FROM #Test t1
JOIN
(SELECT TOP 1 Id,Value,COUNT(Id) Cnt FROM #Test
GROUP BY Id,Value
HAVING COUNT(Id) > 1
ORDER BY Id,Value) t2 ON t2.Id = t1.Id AND t2.Value = t1.Value

SET ROWCOUNT 0
END

when we handling a large database like 10GB size of records.

how can i find duplicates and delete that duplicates.

finding duplicate is not a problem but how to remove those duplicates ?

Please suggest me.

Thank u!