Design a unique ID generator in HLD - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the main purpose of a unique ID generator in system design?

To create identifiers that are unique across the system, ensuring no two entities share the same ID, which helps in tracking, referencing, and managing data reliably.

intermediate

Name two common approaches to generate unique IDs in distributed systems.

1. Using timestamp-based IDs combined with machine identifiers (e.g., Snowflake algorithm). 2. Using Universally Unique Identifiers (UUIDs) which are random or pseudo-random.

beginner

Why is it important to avoid collisions in unique ID generation?

Collisions cause two different entities to have the same ID, leading to data corruption, overwriting, or incorrect data retrieval, which breaks system integrity.

intermediate

What role does scalability play in designing a unique ID generator?

The generator must handle increasing load and number of requests without slowing down or producing duplicates, ensuring consistent performance as the system grows.

advanced

Explain the concept of 'sharding' in the context of unique ID generation.

Sharding means dividing the ID generation responsibility across multiple nodes or machines, each generating IDs in a specific range or with unique prefixes to avoid collisions.

Which of the following is a common method to ensure uniqueness in distributed ID generation?

A Using only random numbers without coordination

B Combining timestamp with machine ID

C Assigning IDs manually by users

D Reusing IDs after deletion

Combining timestamp with machine ID helps create unique IDs across machines and time, reducing collisions.

What is a UUID primarily used for?

A Encrypting data

B Managing user sessions

C Compressing files

D Generating globally unique identifiers

UUIDs are designed to generate unique identifiers that are globally unique without central coordination.

Why might a simple auto-increment ID not be suitable for distributed systems?

A It requires central coordination which can be a bottleneck

B It is too random

C It uses too much memory

D It is not human-readable

Auto-increment IDs need a central source to assign IDs sequentially, which can slow down or fail in distributed setups.

What does 'collision' mean in unique ID generation?

A An ID is not encrypted

B An ID is too long

C Two IDs are the same for different entities

D An ID is generated slowly

Collision means two different entities get assigned the same ID, which causes errors.

Which property is NOT essential for a good unique ID generator?

A Predictability

B Scalability

C Uniqueness

D Low latency

Predictability is usually avoided to prevent security risks; uniqueness, scalability, and low latency are essential.

Describe the key components and flow of a distributed unique ID generator system.

Think about how different parts work together to create unique IDs without conflicts.

Explain the trade-offs between using UUIDs and timestamp-based IDs like Snowflake for unique ID generation.

Consider uniqueness, readability, and system requirements.