CASCADE delete preview in SQL - Time & Space Complexity

Time Complexity: CASCADE delete preview

O(n)

Understanding Time Complexity

When we delete a record that has related records in other tables, the database may delete those related records automatically using CASCADE delete.

We want to understand how the time to delete grows as the number of related records increases.

Scenario Under Consideration

Analyze the time complexity of this SQL snippet using CASCADE delete.

-- Assume tables: Orders and OrderItems
-- OrderItems has a foreign key to Orders with ON DELETE CASCADE

DELETE FROM Orders
WHERE OrderID = 123;

This deletes one order and automatically deletes all its related order items.

Identify Repeating Operations

Look for repeated actions caused by the CASCADE delete.

• Primary operation: Deleting each related record in the child table (OrderItems).
• How many times: Once for each related order item linked to the deleted order.

How Execution Grows With Input

The time to delete grows with the number of related records to remove.

Input Size (n)	Approx. Operations
10 related items	Deleting 1 order + 10 order items
100 related items	Deleting 1 order + 100 order items
1000 related items	Deleting 1 order + 1000 order items

Pattern observation: The total work increases roughly in direct proportion to the number of related records.

Final Time Complexity

Time Complexity: O(n)

This means the time to complete the delete grows linearly with the number of related records that must be deleted.

Common Mistake

[X] Wrong: "Deleting one record is always a quick, constant-time operation regardless of related data."

[OK] Correct: Because CASCADE delete removes all related records, the time depends on how many related records exist, not just the one main record.

Interview Connect

Understanding how CASCADE delete affects performance shows you can think about real database behaviors beyond simple queries.

This skill helps you explain and predict how data changes impact system speed in practical situations.

Self-Check

"What if the foreign key was set to SET NULL instead of CASCADE? How would the time complexity change when deleting a record?"

Practice

1. What does CASCADE DELETE do in a database?

easy

A. Creates a backup of rows before deletion

B. Prevents deletion of parent rows if child rows exist

C. Deletes only the parent row without affecting child rows

D. Automatically deletes related child rows when a parent row is deleted

Solution

1. Step 1: Understand CASCADE DELETE behavior

   CASCADE DELETE means when you delete a parent row, all related child rows linked by foreign keys are also deleted automatically.

2. Step 2: Compare options with this behavior

   Automatically deletes related child rows when a parent row is deleted matches this behavior exactly. Options A, B, and D describe different or incorrect behaviors.

3. Final Answer:

   Automatically deletes related child rows when a parent row is deleted -> Option D

4. Quick Check:

   CASCADE DELETE = auto-remove child rows [OK]

Hint: CASCADE means child rows go away with parent [OK]

Common Mistakes:

• Thinking CASCADE prevents deletion
• Confusing CASCADE with backup
• Assuming child rows stay after parent deletion

2. Which SQL syntax correctly previews rows that will be deleted by CASCADE when deleting from orders table with customer_id = 5?

easy

A. SELECT * FROM customers WHERE id = 5;

B. DELETE FROM order_items WHERE customer_id = 5;

C. SELECT * FROM order_items WHERE order_id IN (SELECT id FROM orders WHERE customer_id = 5);

D. SELECT * FROM orders WHERE customer_id = 5;

Solution

1. Step 1: Identify child rows linked to parent rows

   To preview cascade deletes, select child rows where foreign key matches parent rows to be deleted. Here, order_items link to orders by order_id.

2. Step 2: Use subquery to find parent rows with customer_id = 5

   The subquery selects orders with customer_id = 5. The outer query selects order_items with those order_ids.

3. Final Answer:

   SELECT * FROM order_items WHERE order_id IN (SELECT id FROM orders WHERE customer_id = 5); -> Option C

4. Quick Check:

   Preview child rows with IN + subquery [OK]

Hint: Use IN with subquery to find child rows [OK]

Common Mistakes:

• Trying to delete instead of preview
• Selecting wrong table without join
• Using wrong column in WHERE clause

3. Given tables departments and employees with employees.department_id referencing departments.id with CASCADE DELETE, what rows will this query return?

SELECT * FROM employees WHERE department_id IN (SELECT id FROM departments WHERE name = 'Sales');

medium

A. All employees who work in the Sales department

B. All departments named Sales

C. All employees regardless of department

D. No rows, syntax error

Solution

1. Step 1: Understand the subquery

   The subquery selects department ids where the name is 'Sales'.

2. Step 2: Outer query filters employees by those department ids

   Employees with department_id matching those ids are returned.

3. Final Answer:

   All employees who work in the Sales department -> Option A

4. Quick Check:

   IN subquery filters child rows correctly [OK]

Hint: IN with subquery filters child rows by parent condition [OK]

Common Mistakes:

• Confusing employees with departments
• Assuming syntax error
• Ignoring the subquery effect

4. You wrote this query to preview cascade deletes:

SELECT * FROM order_items WHERE order_id = (SELECT id FROM orders WHERE customer_id = 10);

Why might this query cause an error?

medium

A. Subquery returns multiple rows but '=' expects one value

B. Table order_items does not exist

C. Missing WHERE clause in subquery

D. Incorrect column name order_id

Solution

1. Step 1: Analyze subquery result

   The subquery selects all order ids for customer_id 10, which can be multiple rows.

2. Step 2: Check operator usage

   The outer query uses '=' which expects a single value, causing an error if multiple rows returned.

3. Final Answer:

   Subquery returns multiple rows but '=' expects one value -> Option A

4. Quick Check:

   Use IN for multiple subquery results [OK]

Hint: Use IN, not =, when subquery returns many rows [OK]

Common Mistakes:

• Using '=' with multi-row subquery
• Assuming table or column names are wrong
• Ignoring subquery result count

5. You want to preview all child rows that will be deleted when deleting customers with country = 'USA'. The schema has customers and orders (orders.customer_id references customers.id with CASCADE). Which query correctly previews all orders to be deleted?

hard

A. SELECT * FROM customers WHERE country = 'USA';

B. SELECT * FROM orders WHERE customer_id IN (SELECT id FROM customers WHERE country = 'USA');

C. DELETE FROM orders WHERE customer_id IN (SELECT id FROM customers WHERE country = 'USA');

D. SELECT * FROM orders WHERE country = 'USA';

Solution

1. Step 1: Identify parent rows to delete

   Customers with country = 'USA' are the parent rows to be deleted.

2. Step 2: Find child rows linked to those parents

   Orders with customer_id matching those customers will be deleted by CASCADE.

3. Step 3: Write query to preview child rows

   Use IN with subquery selecting customer ids from customers where country = 'USA'.

4. Final Answer:

   SELECT * FROM orders WHERE customer_id IN (SELECT id FROM customers WHERE country = 'USA'); -> Option B

5. Quick Check:

   Preview child rows with IN + subquery [OK]

Hint: Preview child rows with IN and parent filter subquery [OK]

Common Mistakes:

• Selecting from wrong table
• Using DELETE instead of SELECT
• Filtering child table by parent's column directly