Explaining distributed systems like I'm five
summary
Starts from a concrete interview problem: uploading a 10 gigabytes CSV and normalizing it, then shows a working pattern—write to S3 / Google Cloud Storage / Azure Blob Storage, publish a notification to RabbitMQ or SNS, have consumers normalize and insert into a database, and expose queries via an API. Explains why this decouples concurrency and independent failures, notes clock skew concerns, and mentions an implemented open source search engine called Lira that runs on edge networks.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:24Distributed systems solve problems
- 3:29Ice cream truck analogy
- 5:17Three laws of distributed systems
- 7:02Global clocks challenge
- 8:27Large file uploads solution
- 9:52Key takeaways
- 10:34Closing remarks
related talks
30mDistributed system design
01h 00mThe Secret Life of Distributed Systems
29mFollowing the Data: A Deeper Look into Distributed Applications
23mDesigning Robust and Scalable Distributed Applications
35mAn Ode to 17 Databases in 39 minutes
11mScaling systems, scaling influence: Lessons from distributed architecture
38mHow To Build a Distributed System (And Should You?)
01h 02mAsynchronous Distributed Systems