Efficient reliable microservices
summary
Describes combining JVM business logic and an embedded distributed Cassandra database inside the same process to build stateful microservices that handle hundreds of billions of records and millions of ops/sec. Explains benefits of co-locating business logic and DB, how embedded state affects reliability and availability, and how this approach improved service performance; also evaluates which databases are suitable to embed in microservices and which are not.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 4:07High-load service issues
- 10:32Losses in typical architecture
- 17:40Stateful service as solution
- 21:18Reliability and fault tolerance
- 29:18Implementing stateful service
- 37:00Cache and DB sync
- 43:52Heavy query optimization
- 47:30Full-text search and indexes
- 50:25Deployment and update issues
- 55:10Summary
related talks
01h 14mEfficient and reliable microservices
47mHow to Choose a Right Database for Your Microservices?
40mHow to Build Ultra-fast Java Microservices. Overcome Pitfalls of Data Persistence
40mRethinking Microservice Persistence: Beyond the Database Monolith
51mData management in a Microservices world
30mDatabases in the Microservices World
31mHarness the Power of a Single Database for Microservices
44mHow to Serve Thousands of Databases from One Microservice Without Losing Your Mind