Float32, Float64

Floating point numbers.

Types are equivalent to types of C:

  • Float32float.
  • Float64double.

We recommend that you store data in integer form whenever possible. For example, convert fixed precision numbers to integer values, such as monetary amounts or page load times in milliseconds.

Aliases:

  • Float32FLOAT.
  • Float64DOUBLE.

When creating tables, numeric parameters for floating point numbers can be set (e.g. FLOAT(12), FLOAT(15, 22), DOUBLE(12), DOUBLE(4, 18)), but ClickHouse ignores them.

Using Floating-point Numbers

  • Computations with floating-point numbers might produce a rounding error.
  1. SELECT 1 - 0.9
  1. ┌───────minus(1, 0.9)─┐
  2. 0.09999999999999998
  3. └─────────────────────┘
  • The result of the calculation depends on the calculation method (the processor type and architecture of the computer system).
  • Floating-point calculations might result in numbers such as infinity (Inf) and “not-a-number” (NaN). This should be taken into account when processing the results of calculations.
  • When parsing floating-point numbers from text, the result might not be the nearest machine-representable number.

NaN and Inf

In contrast to standard SQL, ClickHouse supports the following categories of floating-point numbers:

  • Inf – Infinity.
  1. SELECT 0.5 / 0
  1. ┌─divide(0.5, 0)─┐
  2. inf
  3. └────────────────┘
  • -Inf — Negative infinity.
  1. SELECT -0.5 / 0
  1. ┌─divide(-0.5, 0)─┐
  2. -inf
  3. └─────────────────┘
  • NaN — Not a number.
  1. SELECT 0 / 0
  1. ┌─divide(0, 0)─┐
  2. nan
  3. └──────────────┘

See the rules for NaN sorting in the section ORDER BY clause.

Original article