Part 1 of 9 · 1 min read
Why this page comes first after APIs
In this module you will
- Explain what a load balancer does, and why a pool of servers behind one address is more available than one server.
- Tell a layer 4 load balancer from a layer 7 load balancer, and say which job each one suits.
- Name what a reverse proxy adds: TLS termination, caching, compression, hiding the servers.
- Choose a routing rule (round robin, least connections, hashing, two random choices) and explain the weakness of each.
- Work out how many requests fail while a dead server is still in the pool, and how health checks shorten that time.
- Describe DNS balancing and geo balancing, and place a load balancer in a two-layer design.
A single server has a ceiling and dies sometimes. Almost every design in this course puts more than one copy of a service behind a front door. This page explains that door. The numbers are made up to show the idea, unless the text names a book.