UUID (Universally Unique Identifier) is a 128-bit identifier that is unique across both space and time. UUIDs are also known as GUIDs (Globally Unique Identifiers). They are commonly used in software development to uniquely identify information without requiring a central registration authority.

A UUID is represented as 32 hexadecimal digits, displayed in five groups separated by hyphens, in the form 8-4-4-4-12 for a total of 36 characters (32 alphanumeric characters and 4 hyphens).

Example: 550e8400-e29b-41d4-a716-446655440000

UUID has multiple versions, each designed for different use cases. The version number is encoded in the UUID itself (the first character of the third group).

Background: UUID v1 was one of the original UUID versions defined in RFC 4122. It combines a timestamp with the MAC address of the generating machine to create a unique identifier.

How it works: Uses a 60-bit timestamp (100-nanosecond intervals since October 15, 1582), a 14-bit clock sequence, and a 48-bit node ID (typically the MAC address).

Considerations: Exposes the MAC address and creation time, which may be a privacy concern. Not suitable for security-sensitive applications.

Background: UUID v4 is the most widely used version today. It relies entirely on random or pseudo-random numbers, making it simple to implement and privacy-friendly.

How it works: 122 bits are randomly generated, with 6 bits reserved for version and variant information. This provides approximately 5.3 × 10³⁶ possible UUIDs.

Considerations: Not sortable by creation time. Collision probability is extremely low but theoretically possible. Requires a good random number generator.

Background: UUID v5 generates deterministic UUIDs by hashing a namespace and name using SHA-1. Given the same inputs, it always produces the same UUID. (UUID v3 is similar but uses MD5.)

How it works: Concatenates a namespace UUID with a name string, then applies SHA-1 hashing. The first 128 bits of the hash become the UUID (with version and variant bits set).

Considerations: Deterministic nature means the same input always produces the same UUID. Not suitable when uniqueness per generation is required.

Background: UUID v7 is a newer version proposed in the draft RFC for new UUID formats. It addresses the limitations of v1 by using Unix Epoch timestamps, making UUIDs naturally sortable.

How it works: Uses a 48-bit Unix timestamp (milliseconds since 1970-01-01), followed by random bits. The timestamp is placed in the most significant bits, enabling lexicographic sorting.

Considerations: Still a draft standard (as of 2024), but gaining adoption. Provides better database performance than v4 due to sequential nature. Recommended for new projects that need sortable UUIDs.

This generator supports all four versions. UUID v4 is selected by default as it provides random, unique identifiers suitable for most use cases.