111.1K
SJ1. What is Redis?
Redis is an in-memory key-value data store.
Unlike traditional databases that store data on disk, Redis keeps everything primarily in RAM — that’s why it’s extremely fast (microseconds latency).
⸻
2. How Redis Stores Data Internally
Redis uses:
🔹 Hash Table (Dictionary)
At the core, Redis uses a hash table to store keys.
Because of hashing:
• Lookup → O(1)
• Insert → O(1)
• Delete → O(1)
That’s the main reason for high speed.
3. Data Structures Used by Redis
Redis doesn’t just store strings. Internally, it uses optimized structures like SDS , hash table , Skip list, zip list.
4. Why Redis is So Fast
Redis is fast because:
• ✅ In-memory storage (no disk seek time)
• ✅ Single-threaded event loop (no lock contention)
• ✅ Uses efficient data structures
• ✅ Uses I/O multiplexing (epoll/kqueue)
Single-threaded doesn’t mean slow — it means no context switching overhead.
5. How Redis Handles Persistence
Even though Redis is in-memory, it supports durability:
• RDB → Snapshot-based persistence
• AOF (Append Only File) → Logs every write operation
This ensures recovery after crash.
6. How Redis Handles Expiry
Redis stores TTL metadata alongside keys.
It removes expired keys using:
• Lazy deletion (when accessed)
• Active expiration cycle (background cleanup)
#systemdesigninterview #coding #code #google #ai
@sjain.codes
