How Infinite Federation Works

Stand up a federated query layer, a shared catalog, and product-first governance—then let domains ship value.

Core building blocks

  1. Federated Engine: Trino or Starburst as the execution fabric across sources.
  2. Open Table Format: Apache Iceberg for ACID tables and time-travel.
  3. Catalog: Polaris / Nessie / HMS for table & schema registry.
  4. Object Store: S3 / ADLS / MinIO for cost-efficient lake storage.
  5. Metadata Mesh: OpenMetadata (ownership, lineage, docs, tags).
  6. Policy Plane: OPA (ABAC/RBAC) + row/column masking; SSO groups.
  7. Observability: Query events, costs, SLOs; quality via data contracts.
  8. Developer Access: JDBC/ODBC, REST, and optional MCP for agent workflows.

Logical architecture



  

  
    
      
      
    
    
      
    
  

  
  
  Clients: BI · Notebooks · Apps · LLM Agents

  
  
  
  JDBC · ODBC · REST

  
  
  
  TRINO / STARBURST
  resource groups · catalogs per domain

  
  

  
  
  EXTERNAL SOURCES
  (Snowflake · BigQuery · Postgres …)



Operating model (Data Mesh)

Domain ownership

Each domain publishes discoverable, documented products (with SLAs, owners, contracts).

Platform guardrails

Central team runs engine, catalog, policy, and observability; no blocking domain autonomy.

Self-service

Bootstrap templates, connectors, CI/CD, and access requests via groups & tags.

Getting started (30–60–90)

Days 0–30: Foundation

Days 31–60: First data products

Days 61–90: Scale & standardize

Governance & security (pragmatic)

Success measures