DEV Community

Kelly Okere
Kelly Okere

Posted on

Numeric Data Types in SQL: A Comprehensive Guide

In the world of relational databases, numeric data types play a crucial role in storing and manipulating numerical values. SQL (Structured Query Language) provides a wide range of numeric data types to accommodate different ranges and precisions, ensuring accurate representation and efficient storage of numerical data. Whether you're working with whole numbers, decimal values, or floating-point numbers, understanding the available numeric data types and their characteristics is essential for effective database design and management. In this comprehensive guide, we'll explore the various numeric data types in SQL, their definitions, ranges (both signed and unsigned), use cases, example queries, and relevant DBMS-related information.

  1. Integer Data Types: Integer data types are used to store whole numbers without decimal points. SQL offers several integer data types to accommodate different ranges and storage requirements, with both signed and unsigned variants.

a. INT (or INTEGER):
Definition: The INT data type is the most commonly used integer type and is supported by all major database management systems (DBMS). It typically stores 32-bit integers.
Signed Range: The signed range of INT varies across different DBMS, but it's generally from -2,147,483,648 to 2,147,483,647.
Unsigned Range: The unsigned range of INT is from 0 to 4,294,967,295.
Use Case: Suitable for storing integer values that fall within the specified range, such as product quantities, employee IDs, or customer ages.
Example:

CREATE TABLE orders (
    order_id INT UNSIGNED PRIMARY KEY,
    customer_id INT,
    quantity INT
);

INSERT INTO orders (order_id, customer_id, quantity) VALUES (1, 101, 5);

SELECT order_id, quantity FROM orders WHERE customer_id = 101;
Enter fullscreen mode Exit fullscreen mode

DBMS-related Information:

  • In MySQL and PostgreSQL, INT is a 4-byte signed or unsigned integer.
  • In Oracle, INT is an alias for the NUMBER(38) data type, which stores signed integers up to 38 digits.
  • In SQL Server, INT is a 4-byte signed integer, and there is no unsigned variant.

b. SMALLINT:
Definition: The SMALLINT data type is used to store smaller integer values, typically 16-bit integers.
Signed Range: The signed range of SMALLINT is generally from -32,768 to 32,767.
Unsigned Range: The unsigned range of SMALLINT is from 0 to 65,535.
Use Case: Suitable for storing integer values with a smaller range, such as age ranges, small product quantities, or flag values.
Example:

CREATE TABLE students (
    student_id INT PRIMARY KEY,
    age SMALLINT UNSIGNED
);

INSERT INTO students (student_id, age) VALUES (1, 20);

SELECT student_id, age FROM students WHERE age >= 18;
Enter fullscreen mode Exit fullscreen mode

DBMS-related Information:

  • In MySQL and PostgreSQL, SMALLINT is a 2-byte signed or unsigned integer.
  • In Oracle, SMALLINT is not a supported data type, but you can use NUMBER(4) to store small signed integers.
  • In SQL Server, SMALLINT is a 2-byte signed integer, and there is no unsigned variant.

c. BIGINT:
Definition: The BIGINT data type is used to store larger integer values, typically 64-bit integers.
Signed Range: The signed range of BIGINT is generally from -9,223,372,036,854,775,808 to 9,223,372,036,854,775,807.
Unsigned Range: The unsigned range of BIGINT is from 0 to 18,446,744,073,709,551,615.
Use Case: Suitable for storing large integer values, such as population counts, financial transactions, or unique identifiers.
Example:

CREATE TABLE website_stats (
    session_id BIGINT UNSIGNED PRIMARY KEY,
    total_bytes_transferred BIGINT
);

INSERT INTO website_stats (session_id, total_bytes_transferred) VALUES (12345678901234567, 5678901234);

SELECT SUM(total_bytes_transferred) FROM website_stats;
Enter fullscreen mode Exit fullscreen mode

DBMS-related Information:

  • In MySQL and PostgreSQL, BIGINT is an 8-byte signed or unsigned integer.
  • In Oracle, BIGINT is not a supported data type, but you can use NUMBER(19) to store large signed integers.
  • In SQL Server, BIGINT is an 8-byte signed integer, and there is no unsigned variant.
  1. Decimal Data Types: Decimal data types are used to store precise decimal numbers with a fixed number of digits before and after the decimal point. These data types are ideal for financial calculations or any scenario where precise decimal values are required.

a. DECIMAL (or NUMERIC):
Definition: The DECIMAL (or NUMERIC) data type is used to store decimal numbers with a specified precision and scale. The precision represents the total number of digits, and the scale represents the number of digits after the decimal point.
Range: The range of DECIMAL depends on the specified precision and scale, but it generally supports values up to a specified number of digits.
Use Case: Suitable for storing precise decimal values, such as financial data, measurements, or currency values.
Example:

CREATE TABLE products (
    product_id INT PRIMARY KEY,
    price DECIMAL(10, 2)
);

INSERT INTO products (product_id, price) VALUES (1, 19.99);

SELECT product_id, price, price * 1.08 AS price_with_tax FROM products;
Enter fullscreen mode Exit fullscreen mode

DBMS-related Information:

  • In MySQL, DECIMAL and NUMERIC are synonymous and support a maximum precision of 65 digits.
  • In PostgreSQL, NUMERIC is the preferred data type for storing decimal values, with a maximum precision of 1000 digits.
  • In Oracle, DECIMAL is not a supported data type, but you can use NUMBER(p, s) to store decimal values, where p is the precision and s is the scale.
  • In SQL Server, DECIMAL and NUMERIC are synonymous and support a maximum precision of 38 digits.
  1. Floating-Point Data Types: Floating-point data types are used to store approximate decimal numbers with a wide range of values. These data types are suitable for scientific calculations or scenarios where precision is not a critical requirement.

a. FLOAT (or REAL):
Definition: The FLOAT (or REAL) data type is used to store single-precision 32-bit floating-point numbers.
Range: The range of FLOAT varies depending on the DBMS implementation, but it generally supports values from around 1.175494e-38 to 3.402823e+38.
Use Case: Suitable for storing approximate decimal values, such as scientific calculations or measurements where precision is not critical.
Example:

CREATE TABLE scientific_data (
    experiment_id INT PRIMARY KEY,
    measurement FLOAT
);

INSERT INTO scientific_data (experiment_id, measurement) VALUES (1, 3.14159);

SELECT experiment_id, ROUND(measurement, 2) FROM scientific_data;
Enter fullscreen mode Exit fullscreen mode

DBMS-related Information:

  • In MySQL and PostgreSQL, FLOAT is a 4-byte single-precision floating-point number.
  • In Oracle, FLOAT is not a supported data type, but you can use BINARY_FLOAT or BINARY_DOUBLE to store single-precision or double-precision floating-point numbers, respectively.
  • In SQL Server, FLOAT is a 4-byte single-precision floating-point number, and REAL is a synonym for FLOAT.

b. DOUBLE PRECISION:
Definition: The DOUBLE PRECISION data type is used to store double-precision 64-bit floating-point numbers, providing a larger range and higher precision than the FLOAT data type.
Range: The range of DOUBLE PRECISION varies depending on the DBMS implementation, but it generally supports values from around 2.225074e-308 to 1.797693e+308.
Use Case: Suitable for storing approximate decimal values with a higher precision, such as scientific calculations or complex mathematical operations.
Example:

CREATE TABLE astronomical_calculations (
    calculation_id INT PRIMARY KEY,
    distance_in_light_years DOUBLE PRECISION
);

INSERT INTO astronomical_calculations (calculation_id, distance_in_light_years) VALUES (1, 9.4605284e15);

SELECT calculation_id, ROUND(distance_in_light_years, 2) FROM astronomical_calculations;
Enter fullscreen mode Exit fullscreen mode

DBMS-related Information:

  • In MySQL and PostgreSQL, DOUBLE PRECISION is an 8-byte double-precision floating-point number.
  • In Oracle, DOUBLE PRECISION is a synonym for BINARY_DOUBLE, which is an 8-byte double-precision floating-point number.
  • In SQL Server, DOUBLE PRECISION is not a supported data type, but you can use FLOAT to store double-precision floating-point numbers.

Conclusion:
Numeric data types in SQL are essential for storing and manipulating numerical values in databases. By understanding the various integer, decimal, and floating-point data types, their definitions, ranges (both signed and unsigned), use cases, example queries, and DBMS-related information, you can make informed decisions when designing and managing databases. Proper selection and utilization of these data types will contribute to the overall efficiency, data integrity, and reliability of your database applications. It's crucial to consider the range requirements, precision needs, and specific DBMS implementations when choosing the appropriate numeric data type for your data storage and processing needs.

Top comments (0)