Loading…
While this loads — worth knowing
Redis has data structures — sorted sets, counters, pub/sub. Memcached has plain keys and nothing else.
Loading…
While this loads — worth knowing
Redis has data structures — sorted sets, counters, pub/sub. Memcached has plain keys and nothing else.
Scale journeys
Turn the traffic up on a real company's architecture and watch which piece gives way first. Fix it, and move to the next scale — from one server to a billion users. Then build it yourself.
Sharded data, and IDs that never collide
How a photo-sharing app grows from one server to hundreds of millions of users — told through Instagram's public engineering talks, in AWS terms.
Start the journey →Hundreds of microservices, on its own CDN
How a streaming service grows from one datacenter to a quarter of a billion viewers — the story of moving video to the edge, told in AWS terms and based on Netflix's public engineering talks.
Start the journey →Real-time matching, not page views
How a ride-hailing app grows from one city to the whole world — handling a flood of driver locations, matching riders in real time, and pricing on the fly. Based on Uber's public engineering writing, in AWS terms.
Start the journey →Fan-out on write — except for the famous
How a timeline grows from thousands of users to hundreds of millions — and why the same feature needs opposite designs for an ordinary user and a celebrity. Based on Twitter's public engineering talks, in AWS terms.
Start the journey →Millions of live connections, a tiny team
How a messenger grows from thousands of chats to billions of them — where the resource that runs out is not requests per second but millions of connections held open at once. Based on WhatsApp's public engineering talks, in AWS terms.
Start the journey →