Why choose FLOAT type for balance? How does it differ from DECIMAL? What changes if INTEGER is used?

The FLOAT type is a floating-point number that stores approximate floating-point values. It is chosen when performance is important and some error in calculations is acceptable.

DECIMAL (or NUMERIC) is a fixed-precision type that stores exact decimal values. It is ideal for financial data where accuracy is crucial (e.g., balances, monetary amounts) to avoid rounding errors.

Using INTEGER for balances allows only whole numbers, which is not suitable for currencies with cents. To work around this, sometimes INTEGER is used by multiplying the amount by 100 (e.g., storing cents), but this requires additional logic for display and calculations.

In summary:

• FLOAT: fast, but possible rounding errors — not recommended for financial data.
• DECIMAL: precise, but may be slower.
• INTEGER: precise, but limited to whole numbers; can be used to store amounts in the smallest currency units.

Example of using DECIMAL in SQL:

CREATE TABLE accounts (
  id INT PRIMARY KEY,
  balance DECIMAL(15, 2) -- 15 digits total, 2 after the decimal point
);