You're reading from System Design Guide for Software Professionals Build scalable solutions – from fundamental concepts to cracking top tech company interviews

Product type Paperback

Published in Aug 2024

Publisher Packt

ISBN-13 9781805124993

Length 384 pages

Edition 1st Edition

Concepts

Application Development

Authors (2):

Dhirendra Sinha

Tejas Chopra

Preface

1. Part 1: Foundations of System Design FREE CHAPTER

2. Chapter 1: Basics of System Design

3. Chapter 2: Distributed System Attributes

4. Chapter 3: Distributed Systems Theorems and Data Structures

5. Part 2: Core Components of Distributed Systems

6. Chapter 4: Distributed Systems Building Blocks: DNS, Load Balancers, and Application Gateways

7. Chapter 5: Design and Implementation of System Components –Databases and Storage

8. Chapter 6: Distributed Cache

9. Chapter 7: Pub/Sub and Distributed Queues

10. Part 3: System Design in Practice

11. Chapter 8: Design and Implementation of System Components: API, Security, and Metrics

12. Chapter 9: System Design – URL Shortener

13. Chapter 10: System Design – Proximity Service

14. Chapter 11: Designing a Service Like Twitter

15. Chapter 12: Designing a Service Like Instagram

16. Chapter 13: Designing a Service Like Google Docs

17. Chapter 14: Designing a Service Like Netflix

18. Chapter 15: Tips for Interviewees

19. Chapter 16: System Design Cheat Sheet

20. Index

Estimates and calculations

Let’s start by asking and answering some questions:

How many users do we have? 100M
How many Daily Active Users (DAU) do we have? 10% of total users, making 10M/day.
Queries per second to support for daily active users: People usually order food around lunch and dinner time, so let’s assume that orders will be placed for 3 hours during lunch and 3 hours at dinner time. 10M DAU placing orders within 6 hours means that QPS = 10M/6*60*24 ~= 1200 QPS.
What about traffic spikes? I think 5x is a good assumption: 5 * 1200 ~ 60000 QPS.
How many people are browsing restaurants and doing searches? Assuming 10M users doing 5 search queries, then search QPS = ordering QPS *5 = 60000 QPS.
How many restaurants are on the app? Let’s assume 10M restaurants in the world and also let’s assume 1M restaurants are online on our app. Restaurant data doesn’t change often.

OK, now we have the high-level estimates...

The rest of the chapter is locked