Hardware Requirements
This document outlines the recommended hardware requirements for deploying Waldur in different environments.
Deployment Methods
| Deployment Method | Minimum Requirements | Recommended Configuration | Notes |
|---|---|---|---|
| Docker Compose | • 4 vCPU • 12 GB RAM • 20 GB storage |
• 8 vCPU • 16 GB RAM • 40 GB storage |
Single server deployment, fastest to set up |
| Kubernetes (Helm) | See detailed component breakdown below | See detailed component breakdown below | Production-grade, scalable deployment |
Kubernetes Resource Requirements
Namespace Totals
| Requirement Level | CPU | Memory | Storage |
|---|---|---|---|
| Minimal | 8000m (8 vCPU) | 10000Mi (10 GB) | 20Gi for PostgreSQL |
| Recommended | 12000m (12 vCPU) | 14000Mi (14 GB) | 40Gi for PostgreSQL |
Per-Component Requirements
| Component | CPU Requests | CPU Limits | Memory Requests | Memory Limits | Notes |
|---|---|---|---|---|---|
| Waldur Mastermind API | 500m | 1000m | 2000Mi | 4000Mi | Serves API requests, increase for high traffic |
| Waldur Mastermind Worker | 1000m | 2000m | 2000Mi | 4000Mi | Processes background tasks, critical for performance |
| Waldur Mastermind Beat | 250m | 500m | 500Mi | 1000Mi | Schedules periodic tasks |
| Waldur HomePort | 250m | 500m | 500Mi | 1000Mi | Serves web interface |
| PostgreSQL (Single) | 500m | 1000m | 1024Mi | 2048Mi | Main database, persistent storage |
| PostgreSQL (HA) | 1000m per replica | 2000m per replica | 2048Mi per replica | 4096Mi per replica | For high availability (3 replicas recommended) |
| RabbitMQ | 500m | 1000m | 512Mi | 1000Mi | Message broker |
Storage Requirements
| Component | Minimal Size | Recommended Size | Notes |
|---|---|---|---|
| PostgreSQL | 10Gi | 40Gi | Main database storage, grows with user and resource count |
| RabbitMQ | 2Gi | 5Gi | Message queue persistence |
| Backups | 20Gi | 50Gi | Separate storage for database backups |
Scaling Recommendations
| User Scale | API Replicas | Worker Replicas | PostgreSQL Configuration | Additional Notes |
|---|---|---|---|---|
| Small (<100 users) | 1 | 1 | Single instance | Default values sufficient |
| Medium (100-500 users) | 2 | 2 | Single instance with increased resources | Enable HPA for API |
| Large (500+ users) | 3+ | 3+ | HA with 3 replicas | Enable HPA for all components, increase resource limits |
Performance Factors
Consider increasing resources beyond the recommended values if your deployment includes:
- High number of concurrent users (>50 simultaneous active sessions)
- Large number of resources being managed (>1000 total resources)
- Complex marketplace offerings with many components
- Frequent reporting or billing operations
- Integration with multiple external systems
Hardware Recommendations for Production
| Component | vCPU | RAM | Storage | Network |
|---|---|---|---|---|
| Control Plane Nodes | 4 cores | 8 GB | 100 GB SSD | 1 Gbps |
| Worker Nodes | 8 cores | 16 GB | 200 GB SSD | 1 Gbps |
| Database Nodes | 4 cores | 8 GB | 100 GB SSD | 1 Gbps |
| Load Balancer | 2 cores | 4 GB | 20 GB | 1 Gbps |