Skip to content

robustmq/robustmq

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2,434 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RobustMQ Logo

Ask DeepWiki zread Latest Release License GitHub issues GitHub stars Coverage Build Status Rust Version

Next-generation unified communication infrastructure for AI, IoT, and big data

Introduction & VisionFeaturesRoadmapArchitectureQuick StartDocumentationContributingCommunity


⚠️ Development Status RobustMQ is in early development and not production-ready. We are currently in Phase 1 focusing on building a solid architectural foundation. See Roadmap for detailed development plan and timeline.

🌟 Introduction & Vision

RobustMQ is a next-generation unified messaging infrastructure built with Rust for AI, IoT, and data-intensive systems. It is designed to deliver high throughput, predictable latency, and low operational complexity from edge devices to cloud clusters.

RobustMQ Architecture

🎯 Why RobustMQ

  • High-performance core: Rust-native implementation with low latency and low memory overhead.
  • 🔁 Unified protocol access: MQTT + Kafka compatibility in one system, reducing architecture duplication.
  • 🧠 AI-ready data path: Object storage integration and multi-tier cache to reduce data loading bottlenecks.
  • 🌍 Edge-to-cloud consistency: One architecture for edge gateways, regional clusters, and central cloud.

🧭 Vision

Enable data to move freely and efficiently across AI agents, training clusters, IoT devices, and analytics platforms through one unified messaging layer.

🏗️ Workload Fit

  • 🤖 AI workloads: Lightweight topics for agent communication, shared subscription for elastic training consumers.
  • 📡 IoT workloads: MQTT ingestion with Kafka consumption on the same data plane (MQTT in / Kafka out).
  • 📊 Data workloads: Flexible storage modes for balancing throughput, durability, and cost.

🗺️ RobustMQ Development Roadmap

🚀 Long-term Vision

Enable data to flow freely across AI training clusters, millions of Agents, IoT devices, and the cloud — via the optimal path, at the lowest latency, and with minimal cost.

✨ Roadmap

  • Phase 1: Foundation (Completed) — Built a scalable technical architecture with solid, streamlined, and abstraction-friendly code implementation. Established a robust foundation for multi-protocol adaptation, pluggable storage, extensibility, and elasticity.

  • Phase 2: MQTT Broker (Initial Release) — Delivered a stable, high-performance MQTT Broker with MQTT 3.x/5.0 protocol support, optimized for edge deployment with package size under 20MB. Core protocol capabilities are in place and will continue to evolve in future releases.

  • Phase 3: Kafka Protocol & AI Capabilities (Starting) — With the MQTT Broker initially complete, now launching Kafka protocol adaptation and AI capability development. Prioritizing validation of AI training data caching acceleration and million-level lightweight topic feasibility, using AI workloads to drive Kafka protocol implementation; progressively building out full standard Kafka protocol compatibility on this foundation.

✨ Features

  • ⚙️ Unified Messaging Layer: MQTT 3.1/3.1.1/5.0 + Kafka compatibility, enabling MQTT in / Kafka out in one platform.
  • 🚀 Performance by Design: Rust implementation, low memory usage, low latency, and no GC pause behavior.
  • 🧠 AI Data Acceleration: S3/MinIO integration with multi-tier caching (memory/SSD/object storage) to improve data path efficiency.
  • 🤖 Agent-scale Topics: Support for massive lightweight topic counts with isolation and observability per workload.
  • 🔄 Elastic Consumption Model: Shared subscription to scale consumers beyond rigid partition-concurrency coupling.
  • 💾 Flexible Storage Modes: Memory, hybrid, persistent, and tiered storage strategies configurable per topic.
  • 🌐 Edge-to-Cloud Deployment: Consistent runtime model for edge nodes and cloud clusters, with offline buffering + sync.
  • 🛡️ Ops-friendly Architecture: Single-binary deployment, built-in Raft consensus, and simplified operations.

🚀 Quick Start

One-Line Installation

# Install and start RobustMQ
curl -fsSL https://raw.githubusercontent.com/robustmq/robustmq/main/scripts/install.sh | bash
broker-server start

Quick Test

# Test MQTT connection
mqttx pub -h localhost -p 1883 -t "test/topic" -m "Hello RobustMQ!"
mqttx sub -h localhost -p 1883 -t "test/topic"

Web Dashboard

Access http://localhost:8080 for cluster monitoring and management.

Web UI Web UI Cluster

Try Online Demo

📚 For detailed installation and usage guides, see our Documentation

🔧 Development

# Clone and build
git clone https://github.com/robustmq/robustmq.git
cd robustmq
cargo run --package cmd --bin broker-server

# Build packages
make build              # Basic build
make build-full         # With frontend

📚 For detailed build options, see our Build Guide

📚 Documentation

🤝 Contributing

We welcome contributions! Check out our Contribution Guide and Good First Issues.

🌐 Community

🇨🇳 Chinese Community

  • 微信群: Join our WeChat group for Chinese-speaking users

    WeChat Group QR Code
  • 开发者微信: If the group QR code has expired, welcome to follow our official WeChat account!

    WeChat Official Account QR Code

📄 License

RobustMQ is licensed under the Apache License 2.0, which strikes a balance between open collaboration and allowing you to use the software in your projects, whether open source or proprietary.


Built with ❤️ by the RobustMQ team and contributors.