Why caching reduces latency in HLD - Architecture Impact

System Overview - Why caching reduces latency

This system shows how caching helps reduce the time it takes to get data. When a user asks for information, the system first checks a fast storage called cache. If the data is there, it returns quickly. If not, it gets the data from the slower database and saves it in the cache for next time.

Architecture Diagram

User
  |
  v
Load Balancer
  |
  v
API Gateway
  |
  v
Service
  |
  +-----> Cache
  |         |
  |         v
  +------> Database

Components

User

user

Sends requests to get data

Load Balancer

load_balancer

Distributes incoming requests evenly to services

API Gateway

api_gateway

Handles client requests and routes them to the service

Service

service

Processes requests, checks cache, and fetches data

Cache

cache

Stores frequently accessed data for fast retrieval

Database

database

Stores all persistent data, slower to access

Request Flow - 11 Hops

User → Load Balancer

Load Balancer → API Gateway

API Gateway → Service

Service → Cache

Cache → Service

Service → Database

Database → Service

Service → Cache

Service → API Gateway

API Gateway → Load Balancer

Load Balancer → User

Failure Scenario

Component Fails:Cache

Impact:Cache misses increase, causing all requests to hit the slower database, increasing latency

Mitigation:System continues to work by querying the database directly; cache can be restored or replaced without downtime

Architecture Quiz - 3 Questions

Why does the system check the cache before the database?

A Because the cache stores all data permanently

B Because the database is unreliable

C Because cache is faster to access than the database

D Because the user requests the cache directly

Design Principle

Caching reduces latency by storing frequently requested data in a fast-access layer. This avoids repeated slow database queries, improving user experience and system efficiency.