惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

人人都是产品经理
人人都是产品经理
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
月光博客
月光博客
雷峰网
雷峰网
Google DeepMind News
Google DeepMind News
Y
Y Combinator Blog
Microsoft Security Blog
Microsoft Security Blog
M
MIT News - Artificial intelligence
WordPress大学
WordPress大学
MongoDB | Blog
MongoDB | Blog
V
V2EX
博客园 - 【当耐特】
GbyAI
GbyAI
Stack Overflow Blog
Stack Overflow Blog
I
InfoQ
Martin Fowler
Martin Fowler
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
Hugging Face - Blog
Hugging Face - Blog
B
Blog
V
Visual Studio Blog
D
DataBreaches.Net
C
Check Point Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
F
Fortinet All Blogs

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
I Built an Interactive Kafka Playground (Partitions, Keys...
Devanshu Biswas · 2026-06-27 · via DEV Community

Devanshu Biswas

Kafka's pieces are each simple — partitions, keys, consumer groups, offsets, lag — but together they tie people in knots. So I built an interactive playground where you produce and consume and watch it happen, no broker required.

▶ Live demo: https://dev48v.github.io/kafka-playground/
Source (single file, zero deps): https://github.com/dev48v/kafka-playground

The four ideas, made visible

1. Keys decide partitions. Produce a message with a key and it's hashed to a partition — and the same key always goes to the same partition:

key = "user-42"  →  hash("user-42") % partitionCount  →  Partition 0  (every time)

That single fact is why Kafka can only guarantee ordering within a partition, not across the topic. All of user-42's events stay ordered because they're all in one partition. Produce with no key and messages spread round-robin — faster fan-out, no per-key ordering.

2. A consumer group splits the partitions. Every partition is owned by exactly one consumer in the group. Add a consumer and the partitions rebalance across the group — that's how you scale out consumption.

3. More consumers than partitions = idle consumers. Give a 3-partition topic a 4th consumer and it just sits there. Partitions are the unit of parallelism, so your consumer count is capped by your partition count. Seeing the idle consumer light up grey makes this stick instantly.

4. Offsets and lag. Consuming advances the group's committed offset per partition. The gap between the newest message and the committed offset is lag. Produce faster than you consume and watch lag climb; flip on auto-consume and watch it drain.

Why a playground beats a diagram

You can read "partitions are the unit of parallelism" and still be surprised the first time a consumer goes idle. Dragging the consumer count past the partition count and seeing it happen teaches it in one move. Same with key stickiness — produce three messages with the same key, watch them stack in one partition, and the ordering guarantee stops being abstract.

It's one index.html. A deliberately simplified model (round-robin assignment vs Kafka's range/sticky/cooperative assignors; a small key hash vs murmur2) — but the behaviour you see matches real Kafka.

If this made Kafka click, a star helps others find it: https://github.com/dev48v/kafka-playground