Process / pipelineInformation SystemsDistributed Data IntegrationPipeline

Data Federation

Also known as: federated systems, distributed query

OriginatorAmit Sheth and Paul LarsonYear1990Sources3Related methods1

Data federation is an approach to integrating data from heterogeneous, autonomous, distributed databases without requiring centralized storage. Formalized by Sheth and Larson in 1990, federated systems provide a unified interface to query multiple independent sources while preserving their autonomy and existing schemas.

Key highlights

  • Maintains source autonomy while providing unified query interface
  • Eliminates need for data migration and warehouse infrastructure
  • Supports real-time access to up-to-date data from multiple sources

Intuition

This section is available to Pro members. Upgrade to Pro

How it works

This section is available to Pro members. Upgrade to Pro

When to use it

Implement data federation when sources are autonomous and cannot be centralized, when data changes frequently and replication is impractical, or when you need real-time query access across systems. Assumptions include available network connectivity and acceptable latency for cross-system queries.

Strengths & limitations

Strengths
  • Maintains source autonomy while providing unified query interface
  • Eliminates need for data migration and warehouse infrastructure
  • Supports real-time access to up-to-date data from multiple sources
Limitations
  • Cross-source queries may be slower due to network latency and coordination overhead
  • Sources with poor query capabilities limit optimization opportunities
  • Schema changes in sources require manual updates to federation mappings

Common pitfalls

This section is available to Pro members. Upgrade to Pro

Applications

This section is available to Pro members. Upgrade to Pro

Frequently asked

What is the difference between data federation and data warehousing?

Warehousing copies data to a central location (latency but complete control); federation queries sources remotely (no copies but higher latency). Choose federation for real-time access to autonomous sources; warehousing for analytical queries on integrated history.

How do I handle schema conflicts when federating sources?

Create semantic mappings between local schemas and a global schema. Handle naming conflicts (same concept, different names), type mismatches (string vs. integer), and structural differences (relational vs. hierarchical).

How can I improve federated query performance?

Push predicates to sources to minimize data transfer. Use caching for frequently accessed source data. Parallelize queries across sources. Avoid unnecessary joins across sources; perform within-source joins first.

What happens when a source is unavailable?

Federated systems may return partial results from available sources, fail the query, or use cached data depending on requirements. Design queries and applications to handle source unavailability gracefully.

Sources

  1. 1.
    Sheth, A. P., & Larson, P. A. (1990). Federated database systems for managing distributed, heterogeneous, and autonomous databases. ACM Computing Surveys, 22(3), 183-236.
  2. 2.
    Özsu, M. T., & Valduriez, P. (2011). Distributed and parallel database systems. Proceedings of the ACM SIGMOD International Conference on Management of Data, 1657-1668.
  3. 3.
    Garcia-Molina, H., Ullman, J. D., & Widom, J. (2009). Database Systems: The Complete Book (2nd ed.). Pearson Education.

You have read it. What now?

Cite this page

ScholarGate. (2026, June 3). Data Federation. ScholarGate. https://scholargate.app/information-systems/data-federation