Scaling reads: index, replicate, cache, in that order
Most products read far more than they write, and the reliable way to scale those reads is a fixed order of moves (fix the query, add replicas, add a cache, push to the edge), each sized with arithmetic and each carrying a consistency cost you should name.
What you'll be able to do
- Apply the read-scaling moves in order and say why skipping the first one is the most common mistake
- Compute the replicas and cache hit rate a read target needs, including the cold-cache case
- Route reads so a user always sees their own writes despite replica lag
- Handle a hot key with key replication, a short-lived local cache and request coalescing
Before this: caching-and-stampedes, replication-and-consistency
Log in to read the rest of this lesson
Most products read far more than they write, and the reliable way to scale those reads is a fixed order of moves (fix the query, add replicas, add a cache, push to the edge), each sized with arithmetic and each carrying a consistency cost you should name. This walkthrough runs about 22 minutes, with runnable code you can edit and re-run as you read. A new account gets 7 days of Pro free, no card required.
Log in to continueThe first module of every track is free to read on the web: see what's open in System design.