Engineering at Scale
How to handle 100,000 concurrent users
Load balancing, caching, queues, read replicas, connection pooling, autoscaling and load testing: how to build a platform that survives its busiest minute.
Load balancing, caching, queues, read replicas, connection pooling, autoscaling and load testing: how to build a platform that survives its busiest minute.
What layer 4 and layer 7 load balancers can see and do, ALB vs NLB on AWS, load balancing algorithms, and a typical layout for a learning platform.
Set up Nginx proxy_cache, microcache dynamic pages for one second, bypass logged-in users, and stop pile-ups with cache locks and stale-while-updating.