How do I index a database column?

SQL Server 2005 gives you the ability to specify a covering index. This is an index that includes data from other columns at the leaf level, so you don't have to go back to the table to get columns that aren't included in the index keys.

create nonclustered index my_idx on my_table (my_col1 asc, my_col2 asc) include (my_col3);

This is invaluable for a query that has my_col3 in the select list, and my_col1 and my_col2 in the where clause.

For Python PyTables, indexes don't have names, and they are bound to single columns:

tables.columns.column_name.createIndex()

In SQL Server, you can do the following (see CREATE INDEX (Transact-SQL) for the full list of options):

CREATE [ UNIQUE ] [ CLUSTERED | NONCLUSTERED ] INDEX index_name
    ON <object> ( column [ ASC | DESC ] [ ,...n ] )
    [ INCLUDE ( column_name [ ,...n ] ) ]
    [ WHERE <filter_predicate> ]

(ignoring some more advanced options...)

The name of each index must be unique database-wide.

All indexes can have multiple columns, and each column can be ordered in whatever order you want.

Clustered indexes are unique - one per table. They can't have INCLUDEd columns.

Nonclustered indexes are not unique and can have up to 999 per table. They can have included columns and where clauses.

To create indexes, the following stuff can be used:

1. Creates an index on a table. Duplicate values are allowed: CREATE INDEX index_name ON table_name (column_name)

2. Creates a unique index on a table. Duplicate values are not allowed: CREATE UNIQUE INDEX index_name ON table_name (column_name)

3. Clustered Index: CREATE CLUSTERED INDEX CL_ID ON SALES(ID);

4. Non-clustered index: CREATE NONCLUSTERED INDEX NONCI_PC ON SALES(ProductCode);

Refer to http://www.codeproject.com/Articles/190263/Indexes-in-MS-SQL-Server for details.

1. CREATE INDEX name_index ON Employee (Employee_Name)

2. On a multi column: CREATE INDEX name_index ON Employee (Employee_Name, Employee_Age)

We can use the following syntax to create an index.

CREATE INDEX <index_name> ON <table_name>(<column_name>)

If we do not want duplicate values to be allowed then we can add UNIQUE while creating an index as follows.

CREATE UNIQUE INDEX <index_name> ON <table_name>(<column_name>)

We can create an index on multiple columns by giving multiple column names separated by ','.

An index is not always needed for all the databases. For example: The Kognitio, AKA WX2, database engine, doesn't offer a syntax for indexing, as the database engine takes care of it implicitly.

Data goes on via round-robin partitioning and Kognitio WX2 gets data on and off the disk in the simplest possible way.

Since most of the answers are given for SQL databases, I am writing this for NoSQL databases, specifically for MongoDB.

Below is the syntax to create an index in the MongoDB database using the MongoDB shell.

db.collection.createIndex(<key and index type specification>, <options>)

Example - db.collection.createIndex( { name: -1 } )

In the above example, a single key descending index is created on the "name" field.

Keep in mind MongoDB indexes uses a B-tree data structure.

There are multiple types of indexes we can create in MongoDB. For more information, refer to Indexes.