Logo

Documentation

  • Overview
  • FLIP components
  • FLIP workflow
  • Deploy FLIP
    • Deploy the Central Hub
    • Deploy a FLIP node on-prem
    • Deploy a FLIP node in a TRE
  • System administration
  • User guides
  • Working with FLIP apps

Reference

  • API Reference
  • FAQs
  • Glossary
FLIP
  • Deploy FLIP
  • View page source

Deploy FLIP

A FLIP deployment consists of a Central Hub running in AWS and one or more FLIP nodes running inside participating Trusts. Trust-side nodes can run on Trust-managed infrastructure or inside a Trusted Research Environment (TRE). In every model, the trust polls the Central Hub for tasks — all communication is outbound from the trust, no inbound ports are opened on the trust host.

  • Deploy the Central Hub
    • Architecture
    • Prerequisites
    • Required IAM permissions
    • Full-stack deployment
    • Step-by-step deployment
    • Service authentication
    • Applying schema changes
    • FL image compatibility on upgrade
    • Email setup
    • Status check
    • SSH access via SSM
    • Destroy infrastructure
    • Troubleshooting
  • Deploy a FLIP node on-prem
    • Architecture
    • Prerequisites
    • Recommended end-to-end (hybrid) flow
    • Onboarding an on-prem trust (step by step)
    • Trust authentication
    • Network requirements
    • Troubleshooting
  • Deploy a FLIP node in a TRE
    • Overview
    • TRE Operator Responsibilities
    • Architecture
    • Data Flow
    • Prerequisites
    • Use Case 1: Federated Evaluation
    • Use Case 2: Federated Training
    • Differences from On-Premise Deployment
    • UK TRE Federation Landscape
Previous Next

© Copyright 2026, The London AI Centre for Value-Based Healthcare.

Built with Sphinx using a theme provided by Read the Docs.