Ilya Brin - Software Engineer

History is written by its contributors

Consistent Hashing: The Math Behind Load Balancing

Hey there! 👋

Imagine: you have 1000 servers and millions of requests. How do you distribute load evenly so that adding a new server doesn’t require rehashing everything?

Regular hash(key) % servers becomes a disaster when scaling. Add one server and 90% of data needs to be moved!

But there’s an elegant solution: Consistent Hashing. The mathematics that saves you from chaos in distributed systems.

Let’s dive into the algorithm used by Amazon DynamoDB, Cassandra, and Redis Cluster 🚀

Continue reading

NSQ: The Perfect Message Queue - When Redis Isn't Enough)

Hey distributed systems architect! 🚀

Tired of Redis queues losing messages? Fed up with RabbitMQ’s complexity? Need something that just works at scale? Meet NSQ - the message queue that powers Discord, Bitly, and other high-traffic systems.

While others struggle with queue complexity, you’ll learn when NSQ is not just useful, but absolutely perfect for the job.

Continue reading

Notification System: How to Sync Millions of Devices Without Pain

Hey architect! 📱

Need to build a notification system that works like the big players? Where messages arrive instantly on all user devices, and the server doesn’t crash under load?

Let’s break down how to design a notification service that can handle millions of users and sync state across their devices.

Continue reading