Tech

Notes on the internals and practice of backend and distributed systems frameworks and components.

This section collects notes on the internals and real-world practice of common backend and distributed systems frameworks and components, covering RPC, databases, coordination services, caches and message queues. Each note aims to be close to real engineering scenarios, balancing principle derivation with practical takeaways.

Components

Dubbo — A microservice RPC framework: registry, cluster fault tolerance and the Filter chain.
MySQL — A relational database: indexing internals, transactions and locking.
ZooKeeper — A distributed coordination service: the ZAB protocol, leader election, configuration and locks.
Redis — A high-performance in-memory data structure store: internals and cache design.
Kafka — A distributed messaging engine: the storage model and reliability guarantees.

Dubbo

Alibaba’s open-source microservice RPC framework, focused on service registration/discovery, cluster fault tolerance and an extensible architecture.

Kafka

A distributed messaging engine, focused on the storage/consumption model and the engineering guarantees of reliability and exactly-once semantics.

MySQL

A relational database, focused on the internals of indexing, transactions and locking that underpin performance and correctness.

Redis

A high-performance in-memory data structure store, focused on underlying data structure implementations and typical problems and solutions in cache design.

ZooKeeper

A distributed coordination service, focused on the ZAB protocol, the data model, and ZK-based leader election, configuration centers and distributed locks.