Materialized view vs regular view decision in PostgreSQL

Introduction

We use views to save queries for easy reuse. Materialized views save the query results physically, while regular views run the query every time.

When you want fast access to complex query results that don't change often.

When you need always up-to-date data and can afford to run the query each time.

When your query takes a long time and you want to avoid running it repeatedly.

When you want to save storage space and don't want to store query results.

When you want to refresh data manually or on a schedule for performance.

Syntax

PostgreSQL

CREATE VIEW view_name AS
SELECT column1, column2
FROM table_name
WHERE condition;

CREATE MATERIALIZED VIEW mat_view_name AS
SELECT column1, column2
FROM table_name
WHERE condition
WITH NO DATA; -- optional

A regular view does not store data; it runs the query each time you use it.

A materialized view stores the query result physically and needs to be refreshed to update.

Examples

This regular view shows orders from the last 7 days and runs fresh every time.

PostgreSQL

CREATE VIEW recent_orders AS
SELECT order_id, order_date
FROM orders
WHERE order_date > CURRENT_DATE - INTERVAL '7 days';

This materialized view stores total spent by customers but starts empty until refreshed.

PostgreSQL

CREATE MATERIALIZED VIEW top_customers AS
SELECT customer_id, SUM(amount) AS total_spent
FROM sales
GROUP BY customer_id
ORDER BY total_spent DESC
WITH NO DATA;

This command updates the materialized view with current data.

PostgreSQL

REFRESH MATERIALIZED VIEW top_customers;

Sample Program

This example shows creating a regular view and a materialized view on the same data. It inserts some sales, queries both views before and after refreshing the materialized view.

PostgreSQL

CREATE TABLE sales (
  sale_id SERIAL PRIMARY KEY,
  customer_id INT,
  amount NUMERIC,
  sale_date DATE
);

INSERT INTO sales (customer_id, amount, sale_date) VALUES
(1, 100, '2024-06-01'),
(2, 150, '2024-06-02'),
(1, 200, '2024-06-03');

CREATE VIEW sales_view AS
SELECT customer_id, SUM(amount) AS total
FROM sales
GROUP BY customer_id;

CREATE MATERIALIZED VIEW sales_mat_view AS
SELECT customer_id, SUM(amount) AS total
FROM sales
GROUP BY customer_id
WITH NO DATA;

-- Query regular view
SELECT * FROM sales_view ORDER BY customer_id;

-- Query materialized view before refresh
SELECT * FROM sales_mat_view ORDER BY customer_id;

-- Refresh materialized view
REFRESH MATERIALIZED VIEW sales_mat_view;

-- Query materialized view after refresh
SELECT * FROM sales_mat_view ORDER BY customer_id;

Important Notes

Materialized views use storage space because they save data.

Regular views always show current data but can be slower for complex queries.

You must manually refresh materialized views to update their data.

Summary

Regular views run the query every time you use them.

Materialized views save query results and need refreshing to update.

Choose materialized views for speed with less frequent updates, regular views for always fresh data.

Practice

1. Which statement best describes a regular view in PostgreSQL?

easy

A. It stores data physically and requires manual refresh.

B. It cannot be used in SELECT statements.

C. It runs the underlying query every time it is accessed.

D. It automatically updates data without running queries.

Solution

1. Step 1: Understand what a regular view does

   A regular view is a saved SQL query that runs fresh every time you use it.

2. Step 2: Compare with other options

   Storing data physically and requiring manual refresh describes a materialized view; saying it cannot be used in SELECT statements is false because views can be queried directly; automatically updating data without running queries is incorrect for regular views.

3. Final Answer:

   It runs the underlying query every time it is accessed. -> Option C

4. Quick Check:

   Regular view runs query each time [OK]

Hint: Regular views always run queries fresh on access [OK]

Common Mistakes:

• Confusing materialized views with regular views
• Thinking views store data physically
• Believing views auto-refresh data

2. Which of the following is the correct syntax to create a materialized view in PostgreSQL?

easy

A. CREATE MATERIALIZED VIEW my_mat_view AS SELECT * FROM sales;

B. CREATE VIEW my_view AS SELECT * FROM sales;

C. CREATE TABLE my_mat_view AS SELECT * FROM sales;

D. CREATE TEMP VIEW my_mat_view AS SELECT * FROM sales;

Solution

1. Step 1: Identify syntax for materialized view

   The correct syntax uses CREATE MATERIALIZED VIEW followed by the view name and query.

2. Step 2: Eliminate incorrect options

   CREATE VIEW my_view AS SELECT * FROM sales; creates a regular view, CREATE TABLE my_mat_view AS SELECT * FROM sales; creates a table, CREATE TEMP VIEW my_mat_view AS SELECT * FROM sales; creates a temporary view, not materialized.

3. Final Answer:

   CREATE MATERIALIZED VIEW my_mat_view AS SELECT * FROM sales; -> Option A

4. Quick Check:

   Materialized view syntax includes 'MATERIALIZED' keyword [OK]

Hint: Materialized views use 'CREATE MATERIALIZED VIEW' syntax [OK]

Common Mistakes:

• Using CREATE VIEW instead of CREATE MATERIALIZED VIEW
• Confusing tables with materialized views
• Using TEMP keyword incorrectly

3. Given a materialized view created as:
CREATE MATERIALIZED VIEW sales_summary AS SELECT region, SUM(amount) AS total FROM sales GROUP BY region;
What will be the result of:
SELECT * FROM sales_summary;
immediately after creation?

medium

A. The aggregated total sales per region based on current data.

B. An error because materialized views need manual refresh first.

C. Empty result set because no data is stored yet.

D. The detailed sales rows without aggregation.

Solution

1. Step 1: Understand materialized view creation behavior

   When created, a materialized view stores the query result immediately.

2. Step 2: Determine output of SELECT from materialized view

   It returns the aggregated totals per region as per the query at creation time.

3. Final Answer:

   The aggregated total sales per region based on current data. -> Option A

4. Quick Check:

   Materialized view stores query result on creation [OK]

Hint: Materialized views store data on creation, so SELECT returns stored results [OK]

Common Mistakes:

• Thinking materialized views are empty until refreshed
• Expecting detailed rows instead of aggregated results
• Assuming errors occur without manual refresh

4. You created a materialized view but notice it shows outdated data after the underlying table changes. What is the best way to fix this?

medium

A. Use SELECT * FROM materialized_view to refresh automatically.

B. Drop and recreate the materialized view every time.

C. Restart the PostgreSQL server to refresh views.

D. Run REFRESH MATERIALIZED VIEW to update the data.

Solution

1. Step 1: Identify how to update materialized view data

   Materialized views do not auto-update; REFRESH MATERIALIZED VIEW command updates stored data.

2. Step 2: Evaluate other options

   Dropping and recreating the materialized view every time is inefficient, using SELECT * FROM materialized_view does not refresh it, restarting the PostgreSQL server is unrelated.

3. Final Answer:

   Run REFRESH MATERIALIZED VIEW to update the data. -> Option D

4. Quick Check:

   Use REFRESH MATERIALIZED VIEW to update data [OK]

Hint: Use REFRESH MATERIALIZED VIEW to update stored data [OK]

Common Mistakes:

• Expecting SELECT to refresh data automatically
• Dropping and recreating instead of refreshing
• Restarting server to fix view data

5. You have a large sales table updated hourly. You want fast reports that summarize sales by region but can tolerate data up to 1 hour old. Which is the best choice?

hard

A. Use a regular view to always get fresh data.

B. Use a materialized view refreshed every hour for faster queries.

C. Create a table with triggers to update summaries instantly.

D. Use a temporary view refreshed on each query.

Solution

1. Step 1: Analyze requirements for speed and data freshness

   Reports need fast access and can accept data up to 1 hour old.

2. Step 2: Match options to requirements

   Materialized views store results and can be refreshed hourly, balancing speed and freshness. Regular views run slow on large data. Triggers add complexity. Temporary views do not persist.

3. Final Answer:

   Use a materialized view refreshed every hour for faster queries. -> Option B

4. Quick Check:

   Materialized view + hourly refresh = best for speed and acceptable freshness [OK]

Hint: Materialized views with scheduled refresh balance speed and freshness [OK]

Common Mistakes:

• Choosing regular views despite slow queries
• Using triggers unnecessarily complex
• Confusing temporary views with materialized views