name: OpenMetadata FinOps Framework
description: FinOps guidance for OpenMetadata deployments covering both self-hosted open-source and Collate managed SaaS options. OpenMetadata itself is Apache 2.0 licensed and free; costs arise from infrastructure (self-hosted) or subscription fees (Collate SaaS). OpenMetadata also provides a Cost Analysis feature within the platform to track and optimize data infrastructure spend.
version: '1.0'
url: https://open-metadata.org
framework: FinOps Framework 1.0 FOCUS-aligned
phases:
  - phase: Inform
    description: Understand and gain visibility into OpenMetadata-related costs across deployment options.
    activities:
      - name: Deployment Cost Baseline
        description: For self-hosted deployments, catalog infrastructure costs (compute, storage, networking) for the OpenMetadata server, Elasticsearch/OpenSearch index, and MySQL/PostgreSQL database. For Collate SaaS, identify subscription tier and per-asset costs.
        costDrivers:
          - Compute (JVM heap for OpenMetadata server)
          - Elasticsearch or OpenSearch cluster
          - MySQL or PostgreSQL database
          - Ingestion worker compute (Python-based)
          - Networking and egress
          - Storage for metadata and audit logs
      - name: Collate SaaS Cost Visibility
        description: Track Collate SaaS subscription costs by tier (Free, Premium, Enterprise) and monitor asset utilization against plan limits.
        costDrivers:
          - User seat count vs. plan limit
          - Data asset count vs. plan limit
          - Collate AI add-on licensing
      - name: OpenMetadata Cost Analysis Feature
        description: OpenMetadata includes a built-in Cost Analysis feature (Settings > Data Insights > Cost Analysis) that tracks compute costs across data services, enabling visibility into which tables, pipelines, and services drive data infrastructure spend.
        url: https://docs.open-metadata.org/v1.12.x/how-to-guides/data-insights/cost-analysis
  - phase: Optimize
    description: Identify and act on opportunities to reduce OpenMetadata infrastructure and licensing costs.
    activities:
      - name: Right-Size Infrastructure
        description: For self-hosted deployments, match compute and memory allocations to actual ingestion load. Use Kubernetes HPA or auto-scaling groups to handle variable ingestion workloads.
        recommendations:
          - Start with minimum viable cluster sizes and scale based on observed load
          - Use spot/preemptible instances for ingestion workers
          - Archive historical metadata snapshots to cold storage
      - name: Connector Scheduling Optimization
        description: Schedule metadata ingestion connectors during off-peak hours to reduce concurrent compute pressure. Avoid redundant re-ingestion of unchanged assets.
        recommendations:
          - Use incremental ingestion where connectors support it
          - Stagger ingestion schedules across services to flatten compute peaks
          - Disable connectors for deprecated or decommissioned data sources
      - name: Plan Right-Sizing (Collate SaaS)
        description: Monitor seat utilization and data asset counts against Collate plan limits. Downgrade or upgrade plans based on actual usage patterns.
        recommendations:
          - Audit active user seats quarterly
          - Remove inactive users to stay within tier limits
          - Tag data assets by domain to track utilization by business unit
  - phase: Operate
    description: Establish ongoing FinOps practices for sustained cost management of OpenMetadata environments.
    activities:
      - name: Cost Allocation by Domain
        description: Use OpenMetadata Domains to logically group data assets by business unit or team. Map infrastructure costs to domains for internal chargeback or showback reporting.
        recommendations:
          - Assign all data services and assets to a Domain
          - Use Data Insights dashboards to track per-domain asset counts
          - Correlate ingestion worker costs with domain-level ingestion volumes
      - name: Governance Cost Tracking
        description: Track API usage patterns to identify high-frequency consumers of the OpenMetadata API that may drive infrastructure costs. Monitor search query volumes against Elasticsearch cluster costs.
      - name: Open Source vs. SaaS TCO
        description: Periodically compare Total Cost of Ownership between self-hosted (infrastructure + engineering time) and Collate SaaS (subscription + reduced ops overhead) to ensure the optimal deployment model.
        considerations:
          - Engineering hours for deployment, upgrades, and operations
          - Infrastructure costs (compute, storage, networking)
          - Collate SaaS subscription fee
          - Support and incident response overhead
licensing:
  openSource:
    license: Apache 2.0
    cost: Free
    url: https://github.com/open-metadata/OpenMetadata/blob/main/LICENSE
  managedService:
    provider: Collate
    pricingUrl: https://www.getcollate.io/pricing
    tiers:
      - Free (5 users, 500 assets)
      - Premium (25 users, 5,000 assets — contact for pricing)
      - Enterprise (50+ users, 10,000+ assets — custom pricing)
      - Collate AI add-on (custom pricing)