Redis Fundamentals — Beginner Friendly System Design Guide

Why Applications Become Slow

Imagine Swiggy database receives:

10 lakh requests per minute

Every user:

• searches restaurants
• refreshes homepage
• checks orders

If every request directly hits database:

Users
  │
  ▼
Application Server
  │
  ▼
Database Overload ❌

Problems:

• slow APIs
• high latency
• database crashes

This problem is solved using:

Redis

---

What Is Redis?

Redis is:

An extremely fast in-memory database used mainly for caching and real-time systems.

Redis stores data:

• in RAM (memory)
• instead of slow disk storage

This makes Redis:

• incredibly fast

---

Simple Real-Life Analogy

Imagine restaurant kitchen.

Without Redis:

• chef cooks every order from scratch

Slow.

With Redis:

• popular dishes already prepared nearby

Much faster.

Redis works similarly.

---

Why Redis Is Fast

Normal databases:

• read from disk

Disk is slower.

Redis:

• stores data in memory (RAM)

Memory is extremely fast.

---

Speed Comparison

Storage Type	Approximate Speed
Disk Database	Milliseconds
Redis Memory	Microseconds

Huge difference at scale.

---

Real Example — Swiggy Homepage

Millions of users open Swiggy homepage.

Popular restaurant lists rarely change every second.

Instead of:

• querying database repeatedly

Swiggy may cache results in Redis.

---

Redis Cache Flow

User Opens Swiggy
        │
        ▼
Application Server
        │
        ▼
Check Redis Cache
   │               │
Hit              Miss
   │                │
   ▼                ▼
Fast Response    Query Database
                     │
                     ▼
              Store In Redis

---

What Is Cache Hit?

Data found in Redis.

Very fast response.

---

What Is Cache Miss?

Data not found in Redis.

Application queries:

• database

then stores result in Redis.

---

Why Redis Reduces Database Pressure

Without Redis:

Every Request
      │
      ▼
Database Query

Database becomes bottleneck.

---

With Redis

Most Requests
      │
      ▼
Redis Cache
      │
      ▼
Fast Response

Database load reduces massively.

---

Real Example — Instagram Likes Count

Millions repeatedly view:

• likes
• follower counts
• post counts

Instead of:

• recalculating every time

Redis stores counters in memory.

---

Common Redis Use Cases

Use Case	Example
Caching	Product pages
Sessions	User login state
Counters	Likes/views
Queues	Background jobs
Rate limiting	OTP throttling
Leaderboards	Gaming ranks

---

Redis As Session Store

When user logs into app:

• session data stored temporarily

Redis is commonly used.

---

Session Flow

User Logs In
      │
      ▼
Session Created
      │
      ▼
Stored In Redis
      │
      ▼
Future Requests Validated

---

Why Redis Is Great For Sessions

Because sessions need:

• very fast access
• temporary storage
• quick expiry

Redis handles this efficiently.

---

Real Example — PhonePe OTP Rate Limiting

Suppose user requests OTP repeatedly.

Redis may store:

phone_number → OTP count

to prevent abuse.

---

Rate Limiting Flow

User Requests OTP
        │
        ▼
Check Redis Counter
   │               │
Limit OK        Limit Exceeded
   │                │
   ▼                ▼
Send OTP        Block Request

---

Redis Data Structures

Redis supports multiple structures.

---

Common Redis Structures

Structure	Usage
String	Simple cache
Hash	User profiles
List	Queues
Set	Unique items
Sorted Set	Leaderboards

---

Example — User Cache

Key:
user:101

Value:
{
  "name": "Rahul",
  "city": "Jaipur"
}

---

Example — Leaderboard

Gaming app may use:

Sorted Set

for rankings.

---

Leaderboard Flow

Player Scores
      │
      ▼
Redis Sorted Set
      │
      ▼
Top Rankings Returned

---

Redis Expiry (TTL)

Redis supports automatic expiry.

---

Example

OTP expires after 5 minutes

Redis automatically removes expired data.

---

Expiry Flow

Store OTP In Redis
       │
       ▼
TTL = 300 Seconds
       │
       ▼
Auto Delete After Expiry

---

Why Expiry Matters

Not all cached data should live forever.

Examples:

• OTPs
• sessions
• temporary cache

must expire automatically.

---

Redis In Ecommerce

Flipkart may cache:

• popular products
• trending searches
• homepage recommendations

using Redis.

---

Redis + Database Architecture

Users
  │
  ▼
Application Server
  │
  ▼
Redis Cache
  │
 ┌┴──────────┐
 ▼           ▼
Hit         Miss
 │            │
 ▼            ▼
Fast       Database Query
Response

---

Redis Persistence

Although Redis is memory-based:

• it can save data to disk

using:

• snapshots
• append-only logs

---

Why Persistence Matters

Without persistence:

• server restart loses data

Persistence improves:

• recovery
• durability

---

Redis Replication

Redis can replicate data across servers.

---

Replication Flow

Primary Redis
      │
 ┌────┴────┐
 ▼         ▼
Replica1 Replica2

Benefits:

• high availability
• failover
• read scaling

---

Redis Cluster

Large-scale systems use:

• Redis clusters

to distribute data across nodes.

---

Cluster Flow

Application
     │
     ▼
Redis Cluster
 ┌────┼────┐
 ▼    ▼    ▼
Node1 Node2 Node3

---

Real Example — Hotstar IPL

During IPL:

• massive traffic spikes happen

Redis helps cache:

• live scores
• match stats
• user sessions

reducing backend pressure.

---

Redis vs Traditional Database

Redis	Traditional DB
In-memory	Disk-based
Extremely fast	Slower
Temporary data friendly	Persistent storage
Best for caching	Best for durable storage

---

Important Trade-Off

Redis is fast because:

• data stored in RAM

But RAM is:

• expensive
• limited

So Redis usually stores:

• frequently accessed data only

---

What Should Be Stored In Redis?

Good candidates:

• cache
• sessions
• counters
• temporary state
• hot data

---

What Should NOT Be Stored Only In Redis?

Avoid using Redis as:

• permanent source of truth

for critical systems like:

• banking transactions
• long-term records

---

Common Beginner Misconceptions

---

"Redis Replaces Database"

Usually no.

Redis complements databases.

---

"Everything Should Be Cached"

No.

Poor caching:

• wastes memory
• creates stale data problems

---

"Redis Is Only For Big Companies"

Even small apps benefit from Redis.

---

Real Production Architecture

Users
  │
  ▼
CDN
  │
  ▼
Load Balancer
  │
  ▼
Application Servers
  │
  ▼
Redis Cache
  │
  ▼
Primary Database

---

Final Mental Model

Redis =
Ultra-fast temporary storage
used to reduce latency and database load

---

One-Line Interview Definition

Redis is a high-performance in-memory datastore commonly used for caching, sessions, counters, queues, and real-time workloads to reduce latency and offload databases.