Home/ Questions/Q 8818651
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-14T05:09:32+00:00

I’m using SQL Server 2012. What is the difference between indexing a single non

  • 0

I’m using SQL Server 2012. What is the difference between indexing a single non clustered non-key column and putting multiple columns and then indexing?

If I have table with

  • EmployeeID
  • DepartmentID
  • LocationID
  • Name
  • Address

If I create a single index on

  • EmployeeID
  • DepartmentID

instead of just

  • EmployeeID

What is the difference? I created EmployeeID, DepartmentID index, when I was searching any employee with DepartmentID = X, the SQL Management Studio Actual Execution Plan was suggesting that I create an index on DepartmentID even though the table had a non clustered index on EmployeeID, DeparmentID?

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    In Indexes, order matters. You want to have the biggest subset first, followed by subsequently smaller subsets thereafter. So chances are, you will be looking for employees who belong to a certain department more than you will be looking for departments of a specific employee. Said another way, there are likely more employees per department than there are departments per employee — or at the very least, the consequence of searching is likely to first group by departments.

    So if you change your single index from EmployeeID+DepartmentID to DepartmentID+EmployeeID you’ll likely see that suggestion go away.

    More reading: http://weblogs.sqlteam.com/joew/archive/2008/02/13/60510.aspx

    • 0