Self-Hosting Overview

Learn about Traceloop’s deployment options

Traceloop offers two deployment models tailored to meet diverse customer needs while maintaining robust support for generative AI application workflows.

Need help with self-hosting? Schedule a meeting with our team, and we’ll guide you through the process.

Deployment Options

Hybrid Deployment

Perfect for organizations seeking to:

• Maintain control over data storage while outsourcing compute
• Minimize operational overhead
• Comply with data residency requirements
• Leverage Traceloop’s managed services for processing and monitoring

Key Components:

• Processing Pipelines: Managed by Traceloop
• Dashboard: Hosted and managed by Traceloop
• Data Storage: Remains entirely on your infrastructure
• Metadata Storage: Managed by Traceloop
• Monitoring Models: Managed by Traceloop

Full Platform Self-Hosting

Ideal for organizations requiring:

• Complete control over all components
• Air-gapped environments
• Maximum security and compliance
• Custom infrastructure requirements

Key Components:

• Processing Pipelines: On your infrastructure
• Dashboard: Hosted by Traceloop
• Data Storage: Fully within your infrastructure
• Metadata Storage: Fully within your infrastructure
• Monitoring Models: Fully within your infrastructure

Comparison

Feature	Hybrid Deployment	Full Platform
Processing Pipelines	Traceloop-managed	Customer-managed
Dashboard	Traceloop-hosted	Traceloop-hosted
Data Storage	Customer infrastructure	Customer infrastructure
Metadata Storage	Traceloop-managed	Customer-managed
Monitoring Models	Traceloop-managed	Customer-managed
Data in Motion	Handled by Traceloop	Handled entirely by customer
Security	Shared responsibility	Fully customer-controlled
Deployment Complexity	Lower	Higher

Next Steps

Choose your deployment model:

• Get started with Hybrid Deployment
• Set up Full Platform Self-Hosting

Need assistance? We’re here to help: - Schedule a support call - Join our community Slack