That was really informative messages. Thank you very much sharing.

I’ve created NonClustered Index (IDX_x) and COLUMNSTORE INDEX (idx_csi).

Once I run SELECT query with NonClustered Index, the query takes 30 secs to execute and COLUMNSTORE INDEX takes around 2 seconds to execute.

But, if I try to run the NonClustered Index once again, surprisingly it takes only 2-3 seconds to run the query. So my question is once COLUMNSTORE INDEX is created and ran, is there any mechanism which stores the searched data in memory and will be used later if required. Let me know how it works.

Thanks,

Santosh

You have a Table called Employees with following Columns
EmployeeID INT
EmployeeName VARCHAR(256)
Salary MONEY
DeptID SMALLINT

The table has following Indexes created on it
1. Clustered Index on EmployeeID
2. Non Clustered Index on EmployeeName
3. Non Clustered Index on DeptID

The Employees tables has over 10 million rows.
You want to get the count of Employees, for which you write the following query

SELECT COUNT(*) FROM Employees

What will be the likely query plan for above query? Which index will SQL SERVER use in above query? Please give an explanation for your answer.

please help me in detailed manner

regard

CREATE TABLE DEPT(deptid NUMBER(4) PRIMARY KEY, deptname CHAR(10));
CREATE INDEX dept_deptname_idx ON DEPT(deptname);
/* The DEPT table contains 10,000 rows and the .deptname column contains 10 unique values that are evenly distributed. */

Based on the scenario above, which query needs to use the dept_deptname_idx index?

SELECT deptid FROM DEPT WHERE deptid 1234;
SELECT deptid FROM DEPT WHERE deptname = ‘SALES’;
SELECT DISTINCT deptid FROM DEPT;
SELECT deptid FROM DEPT WHERE deptname ‘SALES’;
SELECT deptid FROM DEPT WHERE deptid < 1234;

please answer it and give me the how and why is pocessing

regard

I faced these in interview,but still have confusions, pls let me clear..

* Will Group function ignore null values?
* Declaring a column (for which we gonna create index)as not null or using indexed column in queries often,which will improve the performance of an index?
*its commonly said that,views and synonyms will get dropped after dropping relevant table..but it becomes invalid and not dropped..only constrains,indexes will get dropped..Am i right??

Thanks
Vijayaraghavan @ Chennai