SQL DOUBLE

What is the SQL DOUBLE data type?

DOUBLE (or DOUBLE PRECISION) stores approximate numeric values using double-precision floating-point.

Welcome to the Galaxy, Guardian!
You'll be receiving a confirmation email

Follow us on twitter :)

Oops! Something went wrong while submitting the form.

Compatible dialects for SQL DOUBLE: MySQL, PostgreSQL, SQL Server, Oracle, SQLite, MariaDB, Snowflake, BigQuery

SQL DOUBLE Full Explanation

DOUBLE, formally defined in the SQL standard as DOUBLE PRECISION, is an approximate numeric data type that uses IEEE-754 64-bit floating-point representation. It can hold very large or very small numbers with roughly 15-17 decimal digits of precision and an exponent range of about 10^-308 to 10^308. Because it is approximate, values are rounded to the nearest representable binary fraction, which can introduce small errors in arithmetic and comparisons.Most vendors expose the type under slightly different names:- MySQL, MariaDB: DOUBLE, DOUBLE PRECISION (synonyms)- PostgreSQL: DOUBLE PRECISION- SQL Server: FLOAT(53) (53 signifies 53-bit binary precision)- Oracle: BINARY_DOUBLE- SQLite: REAL (an alias using 8-byte floats)Unlike exact types such as DECIMAL or NUMERIC, DOUBLE should not be used for money, accounting, or other situations where every fraction of a unit must be preserved. It shines in scientific computing, analytics, statistical calculations, machine-learning pipelines, and anywhere a large dynamic range is valued over absolute exactness.DOUBLE occupies 8 bytes on disk and in memory. Arithmetic operations are usually hardware-accelerated, producing fast calculations. Most dialects allow CAST or implicit conversion between DOUBLE and other numeric types, although down-casting can lose precision.

SQL DOUBLE Syntax

column_name DOUBLE;
-- or, in standard form
column_name DOUBLE PRECISION;

SQL DOUBLE Parameters

Example Queries Using SQL DOUBLE

-- 1. Creating a table with DOUBLE values
CREATE TABLE sensor_readings (
  id INT PRIMARY KEY,
  temperature DOUBLE,
  humidity DOUBLE
);

-- 2. Inserting and selecting DOUBLE data
INSERT INTO sensor_readings (id, temperature, humidity)
VALUES (1, 23.456789, 0.458734);

SELECT id,
       temperature * 9/5 + 32 AS temp_fahrenheit
FROM sensor_readings;

-- 3. Casting an INT to DOUBLE for division
SELECT 5 / 2 AS int_division,
       5 / CAST(2 AS DOUBLE) AS float_division;

Expected Output Using SQL DOUBLE

Queries return 8-byte floating-point numbers
Calculations may show rounding (e
g
, 5 / 2 as 2
5 when cast to DOUBLE)

Use Cases with SQL DOUBLE

Storing scientific sensor data where values span many orders of magnitude
Performing statistical aggregates like standard deviation or correlation
Machine learning feature engineering that involves continuous variables
Simulation or modeling workloads where speed outweighs exactness

Common Mistakes with SQL DOUBLE

Using DOUBLE for financial or accounting totals and encountering rounding errors
Expecting exact equality comparisons (0.1 + 0.2 = 0.3 may be false)
Forgetting to CAST integers to DOUBLE, leading to unintended integer division
Assuming all dialects treat DOUBLE and FLOAT identically; precision differs

First Introduced In

SQL-92

Frequently Asked Questions

When should I use SQL DOUBLE instead of DECIMAL?

Use DOUBLE for scientific, statistical, or ML workloads that need a vast numeric range and can tolerate rounding. Use DECIMAL for exact monetary or inventory counts.

How many digits of precision does DOUBLE provide?

Approximately 15-17 decimal digits. Beyond that range values are rounded.

Is DOUBLE the same as FLOAT?

Not exactly. FLOAT is a generic term that each vendor maps to a specific precision. In most systems, FLOAT without length defaults to the same 8-byte format as DOUBLE, but this is not guaranteed.

How do I avoid equality errors with DOUBLE?

Compare with a tolerance (epsilon) instead of exact equality, e.g., WHERE ABS(a - b) < 1e-9;