1. High-Level Design (HLD)

Purpose: Focuses on the overall structure and architecture of the system. It provides a bird’s-eye view of how the system will function and interact with external systems.

Key Characteristics of HLD:

• Scope: Describes the system as a whole and its major components.
• Abstraction: Provides a conceptual overview without diving into implementation details.
• Audience: Primarily targeted at stakeholders, project managers, architects, and senior developers.
• Output:
  • Architecture diagrams
  • Data flow diagrams
  • Component relationships (e.g., services, databases, APIs)
  • Technology stack decisions
• Focus Areas:
  • System Architecture: Defines how major components like databases, load balancers, and APIs will interact.
  • Scalability: Addresses horizontal/vertical scaling and capacity planning.
  • Communication Protocols: Determines whether components communicate via REST APIs, gRPC, etc.
  • High-Level Features: Defines broad functionality without getting into specific implementation details.

Examples of HLD:

• Designing an online video streaming service like Netflix:
  • Components: CDN, authentication service, media storage, recommendation engine.
  • Interaction: Media storage communicates with a CDN; recommendation service connects to user profiles.
  • Database: The decision is to use NoSQL for user data and object storage for videos.

---

2. Low-Level Design (LLD)

Purpose: Focuses on the details of individual components or modules within the system. It breaks down the high-level design into specific implementation steps.

Key Characteristics of LLD:

• Scope: Deals with the inner workings of each component/module.
• Abstraction: Provides a detailed blueprint that includes how the system will be coded.
• Audience: Targeted at developers and engineers responsible for implementation.
• Output:
  • Class diagrams
  • Data structure details
  • Algorithms and logic
  • API contracts (inputs/outputs, error codes)
  • Pseudo-code
• Focus Areas:
  • Implementation Details: Specifies methods, classes, and how logic will be implemented.
  • Data Structures and Algorithms: Defines what will be used for tasks like searching, sorting, or caching.
  • API Specifications: Details input, output, and error handling for individual APIs.
  • Database Queries: Provides SQL or NoSQL query details.

Examples of LLD:

• For the video streaming service:
  • User authentication module:
    • API endpoint: /login
    • Input: Email and password
    • Logic: Validate credentials against the database using hashing.
    • Output: Return JWT token on success, error on failure.
  • Media storage service:
    • Algorithm: Store videos as chunks and map chunks to user data.
    • Data structure: Use hashmaps for chunk indexing.