Cost & Capacity Estimation

Learn back-of-the-envelope estimation: traffic, storage, bandwidth, servers, caches, queues and cloud cost, with worked examples and interview practice.

Learn Cost & Capacity Estimation free, in English or Hindi, with lessons, shorts and interview questions.

Start course → · Interview questions

30 lessons · 10 modules · Free

What you'll learn

  • Convert users and actions into average and peak QPS, using powers of ten and correct units.
  • Estimate storage and bandwidth, including replication, indexes, headroom and egress.
  • Size servers, connection pools, caches, shards and queues with Little's Law and utilisation targets.
  • Explain why latency explodes near saturation and how fan-out amplifies tail latency.
  • Model monthly cloud cost and unit economics, and choose on-demand, committed or spot capacity.
  • Validate estimates with load tests and monitoring, and present an estimate clearly in an interview.

Syllabus

Why Estimate: The Mental Model

  1. Why Engineers Estimate
  2. A Repeatable Estimation Process
  3. Units, Powers of Ten and Bits vs Bytes

Numbers Every Engineer Should Know

  1. Latency Orders of Magnitude
  2. Throughput Rules of Thumb
  3. Time Conversions Without a Calculator

Estimating Traffic

  1. From Users to Requests per Second
  2. Peaks, Spikes and Seasonality
  3. Read/Write Ratio and Fan-out

Storage and Bandwidth

  1. Sizing Stored Data
  2. Replication, Indexes and Headroom
  3. Bandwidth, Ingress and Egress

Compute and Concurrency

  1. Little's Law
  2. How Many Servers?
  3. Utilisation and Queueing Delay

Memory, Caches and Data Stores

  1. Sizing a Cache
  2. Database Capacity: IOPS, Connections, Shards
  3. Queues, Partitions and Backlog

Modelling Cloud Cost

  1. What Actually Drives the Bill
  2. Unit Economics: Cost per User and per Request
  3. On-demand, Committed and Spot Capacity

Capacity Planning in Practice

  1. Validating Estimates with Load Tests
  2. Forecasting Growth and Lead Time
  3. Monitoring Saturation

Pitfalls, Tail Latency and Optimisation

  1. Common Estimation Mistakes
  2. Tail Latency and Fan-out
  3. Optimising Cost Without Breaking Capacity

Revision and Interview Prep

  1. Cheat Sheet: Formulas and Anchors
  2. Worked Interview Problem: A URL Shortener
  3. Mixed Self-Check

Related courses