Principle:Risingwavelabs Risingwave Docker Cluster Deployment
| Knowledge Sources | |
|---|---|
| Domains | Deployment, Infrastructure |
| Last Updated | 2026-02-09 07:00 GMT |
Overview
A containerized deployment mechanism that provisions a complete streaming database cluster with storage, monitoring, and message queue infrastructure using Docker Compose.
Description
Docker Cluster Deployment provides a reproducible way to run RisingWave with all its dependencies. The deployment packages the streaming database alongside its required infrastructure: object storage (MinIO/S3), metadata store (PostgreSQL), monitoring (Prometheus + Grafana), and optionally a message queue (Redpanda/Kafka).
RisingWave offers multiple Docker Compose profiles for different deployment topologies:
- Standalone: All RisingWave components (meta, compute, frontend, compactor) in a single process
- Distributed: Separate containers for each component, allowing independent scaling
- Storage Variants: 10+ storage backends including MinIO, AWS S3, GCS, Azure Blob, HDFS, Alibaba OSS, Huawei OBS
Each profile is self-contained and includes healthchecks, resource limits, and proper startup ordering.
Usage
Use Docker Compose deployment when:
- Developing and testing locally
- Running integration tests
- Deploying in environments without Kubernetes
- Evaluating RisingWave with minimal setup
Theoretical Basis
Container orchestration follows the infrastructure-as-code pattern:
docker-compose.yml
|
+-- postgres-0 (metadata store)
+-- minio-0 (object storage)
+-- risingwave-standalone (database)
| +-- meta-node (port 5690)
| +-- compute-node (port 5688)
| +-- frontend (port 4566)
| +-- compactor (port 6660)
+-- prometheus-0 (metrics)
+-- grafana-0 (dashboards)
+-- message_queue (Redpanda)