Choosing a Deployment Option
Decision guide to help you select the right deployment method for your needs
🚀 I just want to try it out
Simple standalone deployment that runs locally. Doesn't need a lot of planning or resources to be set up.
Option 1: Docker/Podman Compose
The easiest way to get started. Run TrustGraph locally in 15-30 minutes.
Best for:
- First time trying TrustGraph
- Learning and experimentation
- Local development
Requirements:
- 8GB RAM, 4 CPU cores
- Docker or Podman installed
- 20GB disk space
Option 2: Minikube
Run TrustGraph in a local Kubernetes cluster. Great for learning Kubernetes.
Best for:
- Learning Kubernetes
- Testing K8s configurations
- Production-like environment locally
Requirements:
- 16GB RAM, 8 CPU cores
- Minikube and kubectl installed
- 50GB disk space
🇪🇺 I need to use a European Cloud
Deploy on European cloud providers with GDPR compliance and EU data residency. Keep your data within European borders for regulatory compliance and data sovereignty.
Option 1: OVHcloud
Europe's largest cloud provider with Managed Kubernetes and AI Endpoints.
Best for:
- European data sovereignty
- No egress fees
- Multi-region deployments
Key features:
- GDPR native compliance
- 40+ data centers worldwide
- Anti-DDoS included
Option 2: Scaleway
Cost-effective European cloud with Kubernetes Kapsule and Generative AI services.
Best for:
- Budget-conscious deployments
- GDPR compliance
- Open source commitment
Key features:
- EU-based infrastructure
- Competitive pricing
- Mistral AI integration
💡 Data Sovereignty: Both providers ensure your data remains within EU boundaries, meeting strict European data protection regulations including GDPR. This is essential for organizations handling EU citizen data or operating under EU regulatory frameworks.
☁️ I need a global cloud provider
Deploy on major global cloud platforms with enterprise-grade infrastructure, high availability, and comprehensive managed services.
Option 1: AWS RKE
Production-ready RKE2 Kubernetes cluster on AWS with Bedrock AI integration.
Best for:
- AWS-committed organizations
- High availability requirements
- Enterprise production
Key features:
- AWS Bedrock integration
- RKE2 security hardening
- Auto-scaling support
Option 2: Azure AKS
Managed Kubernetes on Azure with AI Foundry and dual AI model support.
Best for:
- Microsoft ecosystem integration
- Enterprise Azure deployments
- Azure Active Directory
Key features:
- Phi-4 and GPT-4o support
- Azure AI Foundry
- Managed Kubernetes
Option 3: Google Cloud Platform
GKE deployment with VertexAI Gemini integration and ML/AI optimization.
Best for:
- ML/AI-focused projects
- VertexAI integration
- Google technology stack
Key features:
- VertexAI Gemini Flash 1.5
- GKE managed Kubernetes
- Free credits available
💡 Enterprise Features: All global cloud providers offer high availability, auto-scaling, enterprise support, and comprehensive managed services. Choose based on your existing cloud commitments and AI service preferences.
🏢 I want to self-host
Deploy on your own infrastructure with complete control over hardware, data, and operations. Perfect for high-performance requirements and maximum data sovereignty.
Option 1: Intel Gaudi
High-performance AI deployment with Intel Gaudi and GPU accelerators for large models.
Best for:
- GPU-accelerated workloads
- Large model inference (70B+)
- High-performance computing
Key features:
- Intel GPU/Gaudi optimization
- Llama 3.3 70B support
- vLLM and TGI servers
💡 Self-Hosting Benefits: Complete control over your data and infrastructure, no vendor lock-in, and the ability to optimize for your specific hardware. Ideal for organizations with strict data governance requirements or specialized performance needs.