Basic Concepts

ACID Principle

• Atomicity: the entire transaction happens at once, or doesn't happen at all
• Consistency: database must be consistent before and after transactions
• Isoluation: multiple transactions can occur independently without inteference
• Durability: changes of a successful transaction occurs even if the system failure occurs

C10K problem?

• the challenge of designing and implementing a server that can efficiently handle a large number of concurrent client connections, specifically 10,000.
• solution:
  • asynchronous I/O
  • Non-Blocking I/O
  • Efficient Data Structures
  • Connection Pooling

Thundering Herd problem?

• when a large number of requests try to access a resources in a small time frame due to reasons like cache eviction, hardware restart, etc.
• Flooded backend may further cause system failures
• To mitigate this issue
  • randomized cache expiration time
  • rate limiting
  • backend sharding
  • asynchronous cache updates
  • backoff time on client side