Indexes

Definition: Indexes are created to speed up the retrieval of data from a database table. By creating an index, you provide a structured way for the database to quickly locate rows based on the values in one or more columns.

Example:

CREATE INDEX idx_customer_name ON Customers(Name);

This statement creates an index named idx_customer_name on the Name column of the Customers table. With this index, the database can quickly locate customers by their name.

Definition: Unique indexes ensure that all the values in the indexed column are unique, meaning no two rows can have the same value in that column. Unique indexes are commonly used to enforce the uniqueness of values in columns like email addresses or user IDs.

Example:

CREATE UNIQUE INDEX idx_unique_customer_email ON Customers(Email);

This statement creates a unique index named idx_unique_customer_email on the Email column of the Customers table. This ensures that no two customers can have the same email address, thereby maintaining data integrity.

Definition: Composite indexes are indexes on multiple columns of a table. They are used when queries often filter or sort by more than one column. Composite indexes can significantly improve the performance of such queries.

Example:

CREATE INDEX idx_customer_name_address ON Customers(Name, Address);

This statement creates a composite index named idx_customer_name_address on the Name and Address columns of the Customers table. This index will speed up queries that filter or sort by both the name and address of customers.

Definition: Dropping indexes is the process of removing an index from the database. This can be necessary when an index is no longer needed, or if it negatively impacts performance due to its maintenance overhead.

Example:

DROP INDEX idx_customer_name;

This statement drops the index named idx_customer_name from the Customers table. Once dropped, the database will no longer use this index to speed up queries on the Name column, and queries on this column might become slower

Let’s consider a practical example where we create various types of indexes on a Customers table and then demonstrate their usage and benefits.

First, let’s create a Customers table to work with:

CREATE TABLE Customers (
    CustomerID int PRIMARY KEY,
    Name varchar(255),
    Email varchar(255),
    Address varchar(255)
);

This statement creates a Customers table with CustomerID, Name, Email, and Address columns. CustomerID is the primary key.

Next, we insert some sample data into the Customers table:

INSERT INTO Customers (CustomerID, Name, Email, Address)
VALUES
(1, 'John Doe', 'john@example.com', '123 Elm Street'),
(2, 'Jane Smith', 'jane@example.com', '456 Oak Street'),
(3, 'Jim Brown', 'jim@example.com', '789 Pine Street');

This statement inserts three rows into the Customers table.

Now, let’s create an index on the Name column to speed up queries that search by name:

CREATE INDEX idx_customer_name ON Customers(Name);

This index will allow the database to quickly locate rows where the Name column matches the search criteria.

With the index created, we can now perform a query to see the benefits:

SELECT * FROM Customers WHERE Name = 'Jane Smith';

The database uses the idx_customer_name index to quickly find the row where the Name is ‘Jane Smith’.

To ensure that no two customers can have the same email address, we create a unique index on the Email column:

CREATE UNIQUE INDEX idx_unique_customer_email ON Customers(Email);

If we try to insert a duplicate email address, the database will prevent it:

INSERT INTO Customers (CustomerID, Name, Email, Address)
VALUES (4, 'Alice Green', 'jane@example.com', '101 Maple Street');

This statement will fail because ‘jane@example.com‘ already exists in the table.

For queries that filter by both name and address, we create a composite index:

CREATE INDEX idx_customer_name_address ON Customers(Name, Address);

This index will speed up queries like:

SELECT * FROM Customers WHERE Name = 'John Doe' AND Address = '123 Elm Street';

If we decide that the idx_customer_name index is no longer needed, we can drop it:

DROP INDEX idx_customer_name;

This removes the index, and queries that previously benefited from the index might run slower.

By understanding and utilizing indexes, you can significantly improve the performance of your SQL queries, making data retrieval operations faster and more efficient. This guide provides a detailed explanation of different types of indexes, how to create them, and their practical applications.