An S3-compatible object storage server written in Swift, designed as an open, Apache 2.0 alternative to MinIO and RustFS.
AutoMQ/automq
A stateless, Kafka-compatible streaming system that stores data on S3 or S3-compatible object storage instead of local disks.
Open-source alternative to:Amazon MSK
Confluent Cloud
- Stars
- 10,917
- Forks
- 797
- Open issues
- 68
- Last commit
- 6 hours ago
What is it?
AutoMQ is a fork of Apache Kafka that replaces its local-disk storage engine with one built on object storage such as Amazon S3 or MinIO. Brokers hold no persistent state, so partitions and traffic can be moved between nodes without copying data, which the project says allows scaling in or out within seconds.
The project targets two cost and operations problems of running Kafka in the cloud: EBS storage charges, cross-AZ traffic fees, and over-provisioning needed to handle peak load. It uses a WAL and caching layers to reduce object storage request costs, and a rack-aware router so clients in different availability zones read partition metadata for their own zone. A built-in auto-balancer moves partitions and traffic between brokers, so a separate Cruise Control deployment is not needed.
AutoMQ is described as 100% Kafka compatible and exports Prometheus and OpenTelemetry metrics. The repository includes Docker Compose files for a single-node and three-node test setup, with Kubernetes and Linux deployment guides linked. It is licensed under Apache 2.0.
Key features
- 01Stores topic data on S3 or S3-compatible object storage instead of local disks.
- 02Brokers are stateless, allowing scaling in or out within seconds.
- 03Built-in auto-balancer redistributes partitions and traffic without Cruise Control.
- 04Rack-aware routing aims to avoid cross-AZ traffic charges on AWS and GCP.
- 05Exports Prometheus and OpenTelemetry metrics.
- 06Claims full compatibility with the Apache Kafka protocol.
Use cases
- Replacing a self-managed Kafka cluster in the cloud to reduce storage and cross-AZ network costs.
- Running event streaming workloads with frequent scaling against variable traffic.
- Building a Kafka-compatible streaming layer on on-premise S3-compatible storage such as MinIO.
- Evaluating a diskless Kafka deployment locally with the provided Docker Compose setup.
Quick start
- 1
Download the single-node Docker Compose file
curl -O https://raw.githubusercontent.com/AutoMQ/automq/refs/tags/1.5.5/docker/docker-compose.yaml - 2
Start AutoMQ and MinIO with Docker Compose
docker compose -f docker-compose.yaml up -d - 3
Run a test Kafka producer against AutoMQ
docker run --network automq_net automqinc/automq:latest /bin/bash -c "/opt/automq/kafka/bin/kafka-producer-perf-test.sh --topic test-topic --num-records=1024000 --throughput 5120 --record-size 1024 --producer-props bootstrap.servers=server1:9092 linger.ms=100 batch.size=524288 buffer.memory=134217728 max.request.size=67108864" - 4
Stop and remove the setup when finished
docker compose -f docker-compose.yaml down
Things to know
- The Quick Start is for evaluation only and is not suitable for production use.
- Requires Docker, at least 4 GB RAM for Docker, and ports 9092 and 9000 free.
- Default latency is hundreds of milliseconds because it writes directly to S3; low latency is in the enterprise version.
Topics
FAQ
- What is automq?
- A stateless, Kafka-compatible streaming system that stores data on S3 or S3-compatible object storage instead of local disks. AutoMQ is a fork of Apache Kafka that replaces its local-disk storage engine with one built on object storage such as Amazon S3 or MinIO. Brokers hold no persistent state, so partitions and traffic can be moved between nodes without copying data, which the project says allows scaling in or out within seconds.
- Is automq free to use? What license does it have?
- automq is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is automq still maintained?
- Actively maintained. The last commit to automq was on Oct 19, 2026.
- What can I use automq for?
- Replacing a self-managed Kafka cluster in the cloud to reduce storage and cross-AZ network costs. Running event streaming workloads with frequent scaling against variable traffic. Building a Kafka-compatible streaming layer on on-premise S3-compatible storage such as MinIO. Evaluating a diskless Kafka deployment locally with the provided Docker Compose setup.
- How do I install automq?
- Download the single-node Docker Compose file: curl -O https://raw.githubusercontent.com/AutoMQ/automq/refs/tags/1.5.5/docker/docker-compose.yaml Start AutoMQ and MinIO with Docker Compose: docker compose -f docker-compose.yaml up -d Run a test Kafka producer against AutoMQ: docker run --network automq_net automqinc/automq:latest /bin/bash -c "/opt/automq/kafka/bin/kafka-producer-perf-test.sh --topic test-topic --num-records=1024000 --throughput 5120 --record-size 1024 --producer-props bootstrap.servers=server1:9092 linger.ms=100 batch.size=524288 buffer.memory=134217728 max.request.size=67108864" Stop and remove the setup when finished: docker compose -f docker-compose.yaml down
- Where can I get automq?
- The source code and installation instructions are on GitHub: https://github.com/AutoMQ/automq
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source, self-hostable backend platform for building apps, offering Auth, Databases, Storage, Functions, Messaging, Realtime, Hosting and MCP access for agents.
Self-hosted audiobook and podcast server with multi-user support, progress syncing, and a web app plus Android and iOS clients.
Apache Druid is a high performance real-time analytics database designed for fast queries and ingestion, often used as an open source data warehouse alternative.
Pipedream is a hosted integration platform for building event-driven automations that connect APIs with pre-built steps or custom Node.js, Python, Go, or Bash code.
Free, self-hosted PostgreSQL backup tool with Point-in-Time Recovery, automated restore verification, and a web UI.