specification: FinOps Framework
specificationVersion: '1.0'
alignedWith:
  framework: FinOps Foundation Framework
  frameworkUrl: https://www.finops.org/framework/
  dataSpec: FOCUS
  dataSpecVersion: '1.3'
  dataSpecUrl: https://focus.finops.org/focus-specification/v1-3/
provider: Federal Housing Finance Agency (FHFA)
providerId: fhfa
publisherName: Federal Housing Finance Agency (FHFA)
serviceCategory: API
created: '2026-06-13'
modified: '2026-06-13'
tags:
  - Housing Finance
  - House Price Index
  - Mortgage
  - Government
  - GSE
  - Federal
  - FinOps
  - Cost Management
  - FOCUS
description: >-
  FinOps framework definition for the FHFA API surface. FHFA is a federal
  government agency that provides all data APIs and datasets at no cost to
  consumers. Direct API usage costs are zero. FinOps considerations center
  on internal engineering and infrastructure costs for building and
  maintaining integrations with FHFA data pipelines, storage of downloaded
  datasets, compute for processing large CSV/XML/JSON files, and operational
  overhead of managing recurring data refresh cycles.
principles:
  - name: Visibility
    description: >-
      Track engineering time and infrastructure costs associated with consuming
      FHFA data, even though the data itself is free. Include storage, compute,
      and pipeline maintenance in cost visibility.
  - name: Allocation
    description: >-
      Allocate costs of FHFA data integration by team, product feature, or
      business unit consuming the housing finance data downstream.
  - name: Optimization
    description: >-
      Since FHFA datasets are updated on monthly or quarterly schedules,
      optimize by caching data locally and avoiding redundant downloads.
      Use incremental processing where possible on large files.
  - name: Accountability
    description: >-
      Establish ownership of FHFA data pipelines and budget for storage and
      compute associated with processing large housing finance datasets.
domains:
  - name: Understand Usage and Cost
    capabilities:
      - Data Ingestion
      - Allocation
      - Reporting and Analytics
      - Anomaly Management
  - name: Quantify Business Value
    capabilities:
      - Planning and Estimating
      - Forecasting
  - name: Optimize Usage and Cost
    capabilities:
      - Architecting for Cloud
      - Workload Optimization
      - Rate Optimization
costs:
  - category: Direct API Cost
    description: Zero — all FHFA data is publicly available at no charge.
    unit: USD
    price: '0.00'
    notes: No API keys, subscriptions, or usage fees exist for FHFA public data.
  - category: Data Storage
    description: >-
      Cost of storing downloaded HPI, NMDB, PUDB, and other FHFA dataset files.
      The HPI master CSV file alone can be tens of megabytes; PUDB zipped files
      can be hundreds of megabytes per annual release.
    unit: USD per GB per month
    price: variable
    notes: Allocate based on total FHFA dataset storage footprint.
  - category: Compute Processing
    description: >-
      CPU and memory costs for parsing and transforming large CSV/XML/JSON
      FHFA dataset files into downstream analytics pipelines or databases.
    unit: USD per compute-hour
    price: variable
    notes: Profile processing jobs for HPI and PUDB file ingestion to size correctly.
  - category: Pipeline Maintenance
    description: >-
      Engineering labor and infrastructure to run recurring data refresh
      jobs aligned with FHFA's monthly and quarterly publication schedules.
    unit: USD per engineering-hour
    price: variable
    notes: Automate refresh pipelines to minimize ongoing manual intervention.
optimization:
  - name: Cache Downloaded Files
    description: >-
      FHFA datasets are updated on fixed monthly or quarterly schedules.
      Cache downloaded files locally or in cloud storage and only fetch
      new data when updates are published to avoid redundant large downloads.
  - name: Monitor Publication Schedules
    description: >-
      Subscribe to FHFA news updates or monitor the FHFA data pages to
      detect new dataset releases and trigger pipeline runs only when
      new data is available.
  - name: Incremental Processing
    description: >-
      Where possible, process only newly added records in HPI or PUDB files
      rather than reprocessing full historical datasets on each refresh cycle.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com