Check the below input data and expected output to get the last three records of a table.

Input Data:

ID          Name
----------- ----------
1           ABC
2           XYZ
3           ERZ
4           HYU
5           BNM
6           WER
7           WER
8           CVC
9           ASD

Expected Output:

ID          Name
----------- ----------
7           WER
8           CVC
9           ASD

Create a sample table:

CREATE TABLE tbl_TestTable (ID INT, Name VARCHAR(10))

INSERT INTO tbl_TestTable 
VALUES (1,'ABC'),(2,'XYZ'),(3,'ERZ')
,(4,'HYU'),(5,'BNM'),(6,'WER')
,(7,'WER'),(8,'CVC'),(9,'ASD')

Solution:

;WITH CTE AS
(
	SELECT TOP (SELECT COUNT(1)-3 FROM tbl_TestTable) ID 
	FROM tbl_TestTable  
)
SELECT *  FROM tbl_TestTable 
WHERE ID NOT IN (SELECT ID FROM CTE)

Please try the different solution for this puzzle and share it via comment...

Check the below input data and expected output for deleting the duplicate data by giving an ID. The table doesn’t have any primary key and writes logic without using ROW_NUMBER().

Input Data:

ID          Name
----------- ----------
1           ABC
1           ABC
1           ABC
3           ERZ
3           ERZ
4           HYU
5           BNM
7           WER
7           WER

Expected Output:

ID          Name
----------- ----------
1           ABC
3           ERZ
3           ERZ
4           HYU
5           BNM
7           WER
7           WER

Create a sample table:

CREATE TABLE tbl_TestTable (ID INT, Name VARCHAR(10))

INSERT INTO tbl_TestTable 
VALUES (1,'ABC'),(1,'ABC'),(3,'ERZ')
,(4,'HYU'),(5,'BNM'),(1,'ABC')
,(7,'WER'),(3,'ERZ'),(7,'WER')

Solution:

;WITH CTE AS
(
	SELECT COUNT(1) AS IDCount, ID FROM tbl_TestTable GROUP BY ID
)
DELETE TOP((SELECT IDCount FROM CTE WHERE ID = 1) - 1) FROM tbl_TestTable  WHERE ID = 1

Please try the different solution for this puzzle and share it via comment...