Question

Step 1 - Understand the problem and establish design scope

Questions to ask

• What are the characteristics of the unique IDs
• For each new record, does ID increment by 1
• Do IDs only contains numerical values
• What is the ID length requirement
• What is the scale of the system

Example requirement

• ID must be unique
• IDs are numerical values only
• IDs fit into 64-bit
• IDs are ordered by date
• Ability to generate over 10,000 unique IDs per second

Step 2 - Propose high-level design and get buy-in

Multi-master replication

• Relies on the auto_increment feature of database, where the ID increase by k (k = number of server in user) each time
• Cons:
  • Hard to scale with multiple data centers
  • IDs do not go up with time across server
  • It does not scale well when a server is added or removed

UUID

• Pros:
  • Scalable
  • No synchronisation issues
• Cons:
  • IDs are 128 bits rather than 64
  • IDs do not go up with time
  • IDs could be non-numeric

Ticket server

• Use auto_increment in a single database server
• Pros:
  • Numeric
  • Easy to implement
• Cons:
  • SOF

Twitter snowflake approach

Step 3 - Design deep dive

• Timestamp ensure the IDs can be ordered by date - 41 bits
  • The maximum timestamp can be represented in 41 bits (in ms) is around 69 years
• Sequence number - 12 bits
  • Field is 0 unless more than one ID is generated per millisecond
  • Maximum 4096 new IDs per ms
• Datacenter ID - 5 bits
  • 32 centers
• Machine ID - 5 bits
• 1 extra bit for future uses, can distinguish positive and negative integers

Step 4 - Wrap up

• Talk about clock synchronisation across timestamps
• Modify the length of each section
• Availability discussion - since this is a mission critical system