DELETE vs TRUNCATE behavior in SQL

Introduction

We use DELETE and TRUNCATE to remove data from tables, but they work differently. Knowing how helps keep your data safe and your database fast.

When you want to remove specific rows from a table based on a condition.

When you want to quickly remove all rows from a table without logging each row deletion.

When you need to keep the table structure but clear its data.

When you want to be able to undo the deletion using a transaction.

When you want to reset the table's identity counter after removing all data.

Syntax

SQL

DELETE FROM table_name WHERE condition;

TRUNCATE TABLE table_name;

DELETE removes rows one by one and can use a WHERE clause to filter rows.

TRUNCATE removes all rows quickly and cannot use a WHERE clause.

Examples

Deletes only employees who work in the Sales department.

SQL

DELETE FROM employees WHERE department = 'Sales';

Removes all rows from the employees table quickly.

SQL

TRUNCATE TABLE employees;

Deletes all rows from orders but logs each deletion and can be rolled back.

SQL

DELETE FROM orders;

Sample Program

This example first deletes the row where name is 'Bob', then shows remaining rows. Next, it truncates the table to remove all rows and shows that the table is empty.

SQL

CREATE TABLE test_table (id INT IDENTITY(1,1), name VARCHAR(20));
INSERT INTO test_table (name) VALUES ('Alice'), ('Bob'), ('Charlie');

-- Delete one row
DELETE FROM test_table WHERE name = 'Bob';
SELECT * FROM test_table;

-- Truncate table
TRUNCATE TABLE test_table;
SELECT * FROM test_table;

Important Notes

DELETE operations can be rolled back if inside a transaction; TRUNCATE usually cannot.

TRUNCATE resets identity counters (auto-increment numbers), DELETE does not.

TRUNCATE is faster for removing all rows because it does not log individual row deletions.

Summary

DELETE removes rows one by one and can filter with WHERE.

TRUNCATE removes all rows quickly without logging each row.

Use DELETE for selective removal and TRUNCATE for fast full table clearing.

Practice

1. Which statement correctly describes the difference between DELETE and TRUNCATE in SQL?

easy

A. TRUNCATE is slower than DELETE because it logs each row removal.

B. TRUNCATE removes rows one by one and can use a WHERE clause; DELETE removes all rows quickly without logging each row.

C. DELETE and TRUNCATE both remove rows one by one and support WHERE clauses.

D. DELETE removes rows one by one and can use a WHERE clause; TRUNCATE removes all rows quickly without logging each row.

Solution

1. Step 1: Understand DELETE behavior

   DELETE removes rows individually and supports filtering with WHERE.

2. Step 2: Understand TRUNCATE behavior

   TRUNCATE removes all rows quickly without logging each row, and does not support WHERE.

3. Final Answer:

   DELETE removes rows one by one and can use a WHERE clause; TRUNCATE removes all rows quickly without logging each row. -> Option D

4. Quick Check:

   DELETE = selective row removal, TRUNCATE = fast full clear [OK]

Hint: DELETE filters rows; TRUNCATE clears all fast [OK]

Common Mistakes:

• Thinking TRUNCATE supports WHERE clause
• Confusing which command logs each row
• Believing DELETE is faster than TRUNCATE

2. Which of the following is the correct syntax to remove all rows from a table named Employees using TRUNCATE?

easy

A. DELETE FROM Employees;

B. TRUNCATE TABLE Employees;

C. TRUNCATE Employees WHERE id > 10;

D. REMOVE ALL FROM Employees;

Solution

1. Step 1: Recall TRUNCATE syntax

   The correct syntax to remove all rows quickly is TRUNCATE TABLE table_name;.

2. Step 2: Check options for correctness

   DELETE FROM Employees; uses DELETE; REMOVE ALL FROM Employees; is invalid SQL; TRUNCATE Employees WHERE id > 10; incorrectly adds WHERE to TRUNCATE; TRUNCATE TABLE Employees; matches the correct syntax.

3. Final Answer:

   TRUNCATE TABLE Employees; -> Option B

4. Quick Check:

   TRUNCATE syntax = TRUNCATE TABLE [OK]

Hint: TRUNCATE always uses TABLE keyword [OK]

Common Mistakes:

• Adding WHERE clause to TRUNCATE
• Using DELETE syntax for TRUNCATE
• Using non-existent REMOVE command

3. Given a table Orders with 1000 rows, what will be the number of rows after executing DELETE FROM Orders WHERE OrderDate < '2023-01-01'; if 300 rows have OrderDate before 2023?

medium

A. 700 rows remain

B. 0 rows remain

C. 1000 rows remain

D. Syntax error

Solution

1. Step 1: Understand DELETE with WHERE clause

   The DELETE command removes only rows matching the condition: OrderDate < '2023-01-01'.

2. Step 2: Calculate remaining rows

   Since 300 rows match the condition, they are deleted. Remaining rows = 1000 - 300 = 700.

3. Final Answer:

   700 rows remain -> Option A

4. Quick Check:

   1000 - 300 = 700 rows left [OK]

Hint: DELETE with WHERE removes only matching rows [OK]

Common Mistakes:

• Assuming DELETE removes all rows without WHERE
• Confusing DELETE with TRUNCATE behavior
• Thinking syntax error due to date format

4. You run the command TRUNCATE TABLE Customers WHERE Country = 'USA'; but get an error. What is the most likely cause?

medium

A. TRUNCATE does not support WHERE clauses.

B. The table Customers does not exist.

C. TRUNCATE requires a WHERE clause.

D. You must use DELETE instead of TRUNCATE for all tables.

Solution

1. Step 1: Check TRUNCATE command rules

   TRUNCATE removes all rows and does not allow filtering with WHERE.

2. Step 2: Identify error cause

   Using WHERE with TRUNCATE causes a syntax error because TRUNCATE does not support WHERE clauses.

3. Final Answer:

   TRUNCATE does not support WHERE clauses. -> Option A

4. Quick Check:

   TRUNCATE = no WHERE allowed [OK]

Hint: TRUNCATE never uses WHERE clause [OK]

Common Mistakes:

• Assuming TRUNCATE supports WHERE
• Thinking table absence causes this error
• Believing DELETE must always replace TRUNCATE

5. You want to quickly remove all rows from a large table Logs but keep the table structure and reset identity counters. Which command should you use and why?

hard

A. Use DELETE FROM Logs WHERE 1=1; because it is faster than TRUNCATE.

B. Use DELETE FROM Logs; because it removes rows and resets identity counters.

C. Use TRUNCATE TABLE Logs; because it removes all rows fast and resets identity counters.

D. Use DROP TABLE Logs; to remove all rows and keep structure.

Solution

1. Step 1: Identify command for fast full table clearing

   TRUNCATE TABLE removes all rows quickly and resets identity counters.

2. Step 2: Compare with DELETE and DROP

   DELETE removes rows slowly and does not reset identity counters; DROP removes table structure.

3. Final Answer:

   Use TRUNCATE TABLE Logs; because it removes all rows fast and resets identity counters. -> Option C

4. Quick Check:

   TRUNCATE = fast clear + reset identity [OK]

Hint: TRUNCATE clears fast and resets identity counters [OK]

Common Mistakes:

• Thinking DELETE resets identity counters
• Using DROP when structure must be kept
• Believing DELETE is faster than TRUNCATE