TSQL Interview Questions – Part 1
These are the following common question that I faced in various TSQL interviews.
… I’ll be updating this post with more questions.
1. Difference between SQL Server 2000 & SQL Server 2005 features, or new features of 2005 vs 2000.
– Ranking functions (ROW_NUMBER, RANK, DENSE_RANK, NTILE)
– Exception handling (TRY-CATCH block)
– CTE (Common Table Expressions)
– PIVOT, UNPOVIT
– CUBE, ROLUP & GROUPING SET
2. What tools do you use for performance tuning?
Query Analyzer, Profiler, Index Wizard, Performance Monitor
3. What is SQL Profiler? What it does? What template do you use?
More Info: http://www.extremeexperts.com/SQL/Articles/TraceTemplate.aspx
4. How can you execute an SQL query from command prompt?
OSQL & SQLCMD
More Info: http://msdn.microsoft.com/en-us/library/ms162773.aspx
5. Difference between DELETE & TRUNCATE statement? Which statement can be Rollbacked?
– With DELETE we can provide conditional WHERE clause to remove/delete specific rows, which is not possible with TRUNCATE.
– TRUNCATE is faster than DELETE as Delete keeps log of each row it deletes in transaction logs, but truncate keeps log of only de-allocated pages in transaction logs.
– Both statements can be rolled backed if provided in a transaction (BEGIN TRANS). If not then none of them can be rollbacked.
– DELETE is DML just like INSERT, UPDATE, but TRANCATE is DDL, just like CREATE, ALTER, DROP
Complete differences on Delete & Truncate: http://wp.me/p12rgl-7
6. What are extended stored procedures? Can you create your own extended stored-proc?
More Info: http://msdn.microsoft.com/en-us/library/ms175200.aspx
7. How can you execute a DOS command from SQL or through SQL query by using xp_cmdshell?
exec xp_cmdshell 'dir c:\*.exe'
8. How will you insert result set of the above proc in a table?
insert into exec xp_cmdshell 'dir c:\*.exe'
9. What are Cursors and their types? What type do you use most and which one is fast?
FORWARD-ONLY, FAST-FORWARD or READ-ONLY cursors.
Fastest to slowest: Dynamic, Static, and Keyset.
More Info: http://msdn.microsoft.com/en-us/library/ms180169.aspx
Check how to use Cursors optimally: http://sqlwithmanoj.com/2011/02/07/avoid-cursors-or-use-them-optimally/
10. Why you should not use a cursor? What are its alternatives?
Alternatives: while loops with temp tables, derived tables, correlated sub-queries, CASE stmt
More Info: http://www.sql-server-performance.com/tips/cursors_p1.aspx
11. Difference between LEFT JOIN with WHERE clause & LEFT JOIN with no WHERE clause.
OUTER LEFT/RIGHT JOIN with WHERE clause can act like an INNER JOIN if not used wisely or logically.
JOINs and its types: http://sqlwithmanoj.com/2009/03/12/sql-server-joins-and-types/
12. How will you migrate an SSIS package from Development to Production environment?
Do not include db connections and file paths in your workflow, instead create configuration files. This will help in deploying the pkg created in DEV server to Testing and finally to the PROD environment.
More Info: http://msdn.microsoft.com/en-us/library/cc966389.aspx
13. Multiple ways to execute a dynamic query.
EXEC sp_executesql, EXECUTE()
More Info: http://sqlwithmanoj.com/2011/03/22/execute-or-exec-vs-sp_executesql/
14. Difference between COALESCE() & ISNULL()
More Info: http://sqlwithmanoj.com/2010/12/23/isnull-vs-coalesce/
16. What should be the ideal combination with IN & UNION (ALL) in terms of performance?
a. SELECT *FROM
IN (SELECT… UNION SELECT…)
b. SELECT * FROM
IN (SELECT… UNION ALL SELECT…)
17. What is an IDENTITY column and its usage in INSERT statements?
IDENTITY column can be used with a tables column to make it auto incremental, or a surrogate key.
Check my all blog posts related to IDENTITY: http://sqlwithmanoj.com/tag/identity/
18. Can you create a Primary key without clustered index?
Yes, you can create a Primary Key with a Non Clustered Index. But by default creation of PK automatically creates a clustered index upon the table, check here: http://sqlwithmanoj.com/2015/05/24/sql-myth-primary-key-pk-always-creates-clustered-index/
19. There are two tables one Master and another Feed table, both with 2 columns: ID & Price. Feed table gets truncated and re-populated every day.
Master Table Feed Table ID, Price ID, Price 1 100 1 200 3 200 2 250 5 300 4 500 6 400 6 750 7 500 7 800
Create a job with an optimal script that will update the Master table by the Feed table.
Use MERGE statement to INSERT any new record form the Feed table, update any existing record in Master table and DELETE existing records in Master table that is not present in Feed table.
20. What are CUBE & ROLLUP sets?
CUBE & ROLLUP are the grouping sets used with GROUP BY clause and are very helpful in creating reports.
More Info: http://sqlwithmanoj.com/tag/cube/
21. What new indexes are introduced in SQL Server 2005 in comparison to 2000?
More Info: http://msdn.microsoft.com/en-us/library/ms175049.aspx
22. What are XML indexes, what is their use?
More Info: http://msdn.microsoft.com/en-us/library/ms345121%28SQL.90%29.aspx
23. How many types of functions (UDF) are there in SQL Server? What are inline functions?
– Scalar functions
– Inline Table-valued functions
– Multi-statement Table-valued functions
More Info: http://sqlwithmanoj.com/2010/12/11/udf-user-defined-functions/
24. How will you handle exceptions in SQL Server programming?
By using TRY-CATCH constructs, putting our SQL statements/scripts inside the TRY block and error handling in the CATCH block, http://sqlwithmanoj.com/2010/06/16/try-catch-exception-handling/.
25. How would you send an e-mail form SQL Server?
Configure Database mail here: http://sqlwithmanoj.com/2010/09/29/database-mail-setup-sql-server-2005/.
… more questions on next post [Part-2].