Search & Indexing at Scale

Load balancing at each layer, consistent hashing, rate limiting, and getting session state out of the way so a service can be cloned. Then search, which is the one subsystem that scales differently from everything else.

This module is 05.5 of the 5 in Scaling & Traffic. It runs about 4 hours across 5 lessons, and closes with Inverted — Serve typeahead over 200 million documents in under 50 ms at the ninety-ninth percentile.