Tutorials

A gossip protocol is the mechanism Cassandra, Consul, and Redis Cluster use so thousands of nodes know who’s alive, without relying on a central coordinator. Here we explain how it works, with Python code and a real Consul cluster.

Two-phase commit is the protocol distributed databases use to confirm a transaction across multiple nodes at once, without leaving data half-written. This guide shows how to implement it with PostgreSQL, its known failure modes, and when it makes sense to replace it with Sagas or distributed consensus.

Every time you type a URL, your device triggers a chain of DNS queries across several servers before it gets an IP address. We show what happens at each step, what record types exist, and how to diagnose resolution issues with dig and nslookup.

A trie is a tree that indexes words letter by letter instead of using a full hash. Here’s how the structure behind instant autocomplete, spell checking, and IP routing works internally, with code examples in JavaScript and Python.

Generators in Python let you iterate over large sequences without loading them fully into memory. This article explains how yield works, when to use it, and how it differs from a list or a regular function.

HTTP/2 replaced HTTP/1.1’s multiple TCP connections with a single connection multiplexed and compressed with HPACK. This guide explains how it works internally, how to configure it in Nginx and Node.js, and when it actually improves a site’s performance.