Storage: 0's and 1's stored in a useful structure

Foundations of Practical System Design

System Design Course for Junior Engineers

Kay Ashaolu

What is Storage?

• Storage is data (0's and 1's) organized in a particular way to save information
• This "particular way" is typically described by one or more data structures

Why Care About Data Structures?

• Raw binary data is not inherently meaningful by itself
• Storage using data structures provides shape and context to these bits
• Efficient organization improves:
  • Retrieval speed
  • Memory usage
  • Ability to scale complex systems

First Category: Lists/Arrays/Vectors

• A linear sequence of data
• Elements are stored in a specific order
• Often indexed by integers starting from 0
• Can be extended into multiple dimensions (2D, 3D, etc.)

Arrays and Multidimensional Structures

• 1D arrays: Simple linear lists (e.g., [a, b, c, d])
• 2D arrays: Think of spreadsheets or tables (rows and columns)
• Higher dimensions: Useful in advanced applications (e.g., matrices in math, tensors in machine learning)

Advantages of Arrays

• Easy indexing: Access elements by their position
• Often memory-efficient: Contiguous allocation can improve cache performance
• Ideal for scenarios where order matters or when random access is frequent

Second Category: Dictionaries/Objects

• Key-value pairs, like a real dictionary word lookup
• Instead of numeric indices, you define the keys
• Flexible: You can attach descriptive names to data elements

Dictionaries for Real-World Models

• Represent complex entities naturally:
  • Example: A “person” object with keys like name, age, height
• No need to remember integer indices
• Extend objects with attributes and related data easily

Third Category: Trees and Graphs

• Data points connected by relationships rather than positions or keys
• Trees: Hierarchical structure (like a family tree)
• Graphs: More general connections (networks, social graphs, linked documents)

Power of Trees and Graphs

• Model relationships and dependencies
• Efficiently represent connected structures (e.g., routes on a map)
• Enable complex queries based on connectivity rather than just order or keys

Comparing the Three Categories

• Arrays/Lists: Ordered sequences, fast indexing, simple structure
• Dictionaries/Objects: Keyed lookups, descriptive data modeling
• Trees/Graphs: Relationship-based, complex connectivity, flexible representations

The Common Thread

• All structures store and manipulate the same basic binary data
• Choosing the right structure depends on:
  • The type of data you have
  • The operations you need to perform (lookup, insert, traverse)
  • The patterns of data relationships

Efficiency and Scale

• Proper data structures ensure:
  • Faster responses at scale
  • Better use of memory
  • More intuitive code and system design

Gaining Intuition

• Recognize the patterns in data storage
• Understand how arrays, dictionaries, and graphs influence system architecture
• This intuition helps you design robust, maintainable, and efficient systems

Next Steps

• In future lessons, we’ll see how to use these structures with tasks
• We’ll explore how combining data storage with well-chosen tasks leads to building blocks
• Keep thinking about how data is stored and why it matters!