Generate UUIDs for applications, databases, and APIs

UUID Generator — Cryptographically Random Identifiers for Distributed Systems

UUID (Universally Unique Identifier) generation is deceptively simple and critically important in distributed systems. A UUID is a 128-bit value formatted as a 32-character hexadecimal string separated by hyphens — 550e8400-e29b-41d4-a716-446655440000 — designed to be generated independently on any machine in any location without coordination and still produce a value virtually guaranteed to be unique across all generated values globally. The probability of generating a duplicate UUID v4 (random) with a standard cryptographically random source is 1 in 2^122 — effectively zero for any practical application. Sequential integer IDs generated by a database's auto-increment, by contrast, require coordination through the database that creates a bottleneck in distributed writes, exposes record counts through enumerable IDs, and fails catastrophically when merging records from separate databases. UUID adoption is the architectural decision that removes these constraints.

UUID versions are not interchangeable — each version encodes different information and has different properties that make it suitable for different use cases. UUID v1 encodes the generating machine's MAC address and the current timestamp, producing sortable UUIDs whose temporal ordering matches insertion order but revealing both the machine identity and creation time to anyone who receives the ID. UUID v4 uses purely random bits, revealing no information about generation context but providing no sortability guarantee. UUID v7 (a recent standard) encodes a Unix timestamp in the first 48 bits followed by random bits, combining the sortability of v1 with the privacy of v4 — this makes v7 ideal for database primary keys where index efficiency from sortable IDs matters. The generator produces any version with explicit explanation of the properties and trade-offs so the correct version is selected for the specific system design.

Namespace UUIDs (UUID v3 and v5) generate deterministic identifiers from a combination of a namespace UUID and an arbitrary name string. The same namespace UUID plus the same name string always produces the same output UUID. This determinism is the opposite of randomness — and is precisely what makes namespace UUIDs valuable for systems that need to generate stable, repeatable identifiers for entities that may be encountered multiple times across distributed processes. A data pipeline that ingests records from multiple sources can assign stable UUIDs to each entity by hashing the entity's natural key (email address, product SKU, external ID) against a fixed namespace UUID — producing the same identifier every time the same entity is encountered, enabling idempotent record processing without a centralized ID registry.