Contents
01. Introduction
A Federated Database System (FDBS) is a meta-database management system that transparently maps multiple autonomous database systems into a single federated database. Unlike a monolithic distributed database, the constituent databases in a federation are heterogeneous and retain local autonomy while participating in the federation.
Balancing global integration requirements with local database autonomy and heterogeneous data models.
02. Architecture: Mediator-Wrapper Pattern
Modern federated systems overwhelmingly employ the Mediator-Wrapper architectural pattern to abstract away underlying component database complexities.
03. Handling Schema Heterogeneity
Before a query can be executed across sources, structural conflicts must be resolved. The textbook Global-as-View (GAV) and Local-as-View (LAV) approaches provide mathematical models for integration.
| Approach | Mechanism | Pros & Cons |
|---|---|---|
| Global-as-View (GAV) | Global schema is defined as a view over the local schemas. | Easier query processing, hard to add new dynamic sources. |
| Local-as-View (LAV) | Local sources are defined as views over the global schema. | Easy to add new sources, complex query rewriting required. |
04. Distributed Query Processing
Querying an FDBS involves multiple phases. The execution engine must cost queries not just on CPU/Disk, but on network transfer costs between distributed nodes.
SELECT p.name, m.diagnosis FROM GlobalRegistry_PgSQL p JOIN RegionalHospital_MongoDB m ON p.id = m.patient_id; -- Mediator Execution Plan: 1. Semi-join filter sent to PostgreSQL (Wrapper A) 2. Hash patient IDs in memory 3. Send IN clause payload to MongoDB (Wrapper B) 4. Hash-join results at Mediator node
Full Research Report
The complete LaTeX-typeset document includes deep dives into 2PC consistency models, CAP theorem implications, security, and case studies.
Download PDF Document