Course contentsShow
← System Design
Lesson 1552 of 1,91938. Real-World System: PastebinFree lesson

Initial Architecture Diagram

Mapping out load balancer, app servers, metadata database, object storage, and cache layer relationships.

Initial Architecture Diagram

What you'll learn: How to visually organize Pastebin's core components—load balancer, app servers, databases, object storage, and cache—into a coherent, scalable architecture.

Why Architecture Diagrams Matter

Before writing code, you need a map. An initial architecture diagram shows how data flows through your system, where bottlenecks might occur, and how components communicate. For Pastebin, this means connecting all the pieces you've designed into one cohesive picture.

Mapping the Components

The Request Entry Point

Your load balancer sits at the front, distributing incoming HTTP requests (paste creation and retrieval) across multiple application servers. Think of it as a traffic cop directing cars to different toll booths.

Application Tier

App servers handle business logic: validating input, generating short keys, and coordinating between storage layers. They're stateless, so any server can handle any request—perfect for horizontal scaling.

Data Layer Relationships

  • Metadata database stores small, indexed records (paste ID, creation time, expiration, user ID)
  • Object storage (like S3) holds the actual paste content—your large, blob data
  • App servers query the database for metadata first, then fetch content from object storage only when needed

Cache Layer

The cache sits between app servers and both storage layers. Popular paste metadata and content live here for microsecond-fast retrieval, reducing database hits by 80-90%.

The Flow

  1. User request → Load Balancer
  2. Load Balancer → App Server
  3. App Server checks Cache → if miss, query Database + Object Storage
  4. Response flows back through the chain

Key Takeaway: Your initial architecture diagram connects load balancer, stateless app servers, metadata database, object storage, and cache into a layered design where each component has a clear responsibility and data flows logically from user to storage and back.