---
title: "Data Vault 2.0 Concepts"
canonical: "https://docs.vaultspeed.com/space/VPD/3011870744/Data%20Vault%202.0%20Concepts"
format: markdown
---
### **Why Data Vault 2.0? Addressing Modern Data Challenges**

Organizations today are facing unprecedented data challenges:

- Ever-increasing **data volume, variety, and velocity**
- Constant **changes in source systems** and data structures
- Mounting pressure to deliver **auditable, governed, and reliable** data
- The demand for **real-time, self-service analytics**


**Data Vault 2.0 (DV2.0)** is a business intelligence system that addresses these challenges through a methodology that emphasizes agility, automation, traceability, and scalability. Introduced by Dan Linstedt, it extends the original Data Vault 1.0 by incorporating not just a modeling technique but a complete **architecture, methodology, and implementation framework**.


DV2.0 integrates well with modern data platforms including cloud, NoSQL, and real-time systems. It’s designed to support agile delivery, audit compliance, and long-term maintainability of enterprise data warehouses (EDWs).


**Key Pillars of Data Vault 2.0:**

- **Modeling**: Using Hubs, Links, and Satellites to structure data
- **Architecture**: Layered design supporting raw, business, and information delivery zones
- **Methodology**: Agile, sprint-based delivery cycles with automation
- **Implementation**: Metadata-driven generation of code and models


### **Core Modeling Components – Hubs, Links, Satellites**

DV2.0 relies on a highly modular and extensible data model:

1. **Hubs**

- Represent unique business entities (e.g., Customer, Product)
- Contain **business keys**, a surrogate (hash) key, load date, and record source
- De-duplicated across all source systems
- Serve as the anchor for linking descriptive and relational data

2. **Links**

- Represent **relationships between Hubs** (e.g., Order connects Customer and Product)
- Store the surrogate keys of the participating Hubs plus load metadata
- Support **historical tracking of relationships** and transactional behaviors

3. **Satellites**

- Store **descriptive and historical context** for Hubs and Links
- Fully historized with start and end dates, load date, and source
- Designed for **temporal analytics**, audit trails, and time-travel querying
- Satellites can be split by volatility (frequently vs. rarely changing attributes), domain (finance vs. sales), or classification (PII vs. non-PII)


Additional Constructs:

- **Multi-active Satellites**: Handle multiple valid versions (e.g., multiple addresses)
- **Non-historized Satellites**: Store current state data for operational needs


This model supports **change-resilient design**: new source systems or attribute changes only affect the impacted Satellites or Links, preserving historical data and minimizing regression.


### **Architecture Layers and Automation**

**Data Vault 2.0 architecture** follows a layered approach, supporting end-to-end data lifecycle from raw ingestion to analytics:

#### **1. Ingestion and Pre-Staging Layer**

- Raw data can land in a database struture or can lands in a native format (e.g., CSV, JSON, XML, JDBC)
- In the case of native formats, Pre-staging applies **hard business rules** (type casting, formatting, null handling)
- This layer is used for metadata harvesting and CDC setup (timestamp, log-based, or full compare)

#### **2. Raw Data Vault (RDV)**

- Core of DV2.0: Hubs, Links, Satellites
- Stores raw, immutable, fully historized business data
- No transformations beyond hard rules
- Insert-only: ensures **lineage, auditability, and data preservation**

#### **3. Business Vault (BV)**

- Applies **soft business rules**, transformations, and logic
- Supports derived structures like:
  
  - **Point-in-Time (PIT) tables** for snapshot-style joins
  - **Bridge tables** for hierarchical or many-to-many navigation
  - **Calculated Satellites** (e.g., revenue buckets, classifications)
  
- Rebuildable from RDV, enabling version control and sandbox experimentation


#### **4. Information Marts**

- Dimensional models, star schemas, or wide tables for reporting and analytics
- Serve dashboards, APIs, or machine learning models
- Built from BV or RDV depending on need


### **Value Delivered and Practical Benefits**


**Scalability Across Domains**

- Onboards new sources without rework
- Handles structured, semi-structured, and unstructured data
- Enables decentralized teams to build on a centralized foundation


**Auditability and Compliance**

- Each record tracks load time, source system, and change history
- Ensures **regulatory traceability** (e.g., GDPR, HIPAA, SOX)
- Enables data quality monitoring and lineage reporting


**Agile and Sustainable Development**

- Designed for sprint-based delivery
- Insert-only logic enables parallel work without conflicts
- Supports robust delta deployments and versioning


**Future-Proofing**

- Works across platforms: Snowflake, BigQuery, Synapse, Databricks
- Integrates with modern data stacks (dbt, GitOps, Airflow)
- Enables self-service BI through stable, governed data foundations


**Use Case Snapshots by Industry**

1. **Logistics & Transportation**

A logistics company integrates customer, shipment, and tracking data from SAP, Salesforce, and IoT sensors. With DV2.0:

- Hubs unify customers, locations, and parcels
- Links track relationships like shipments and deliveries
- Satellites store address changes, package statuses, and delay reasons
- PITs and bridges support SLA monitoring and route optimization
- Dashboards deliver real-time logistics insights with full traceability


2. **Healthcare**

A hospital network aggregates patient records, lab results, and insurance claims from EMRs, labs, and billing systems:

- Hubs define patients, physicians, visits, and diagnoses
- Links capture patient-visit-lab relationships
- Satellites track diagnosis history, medication dosages, and procedure outcomes
- Business Vault calculates risk scores and readmission rates
- Enables compliance with HIPAA and supports analytics for patient outcomes


3. **Financial Services**

A bank consolidates customer, transaction, and compliance data across multiple systems:

- Hubs define customers, accounts, and transactions
- Links represent account ownership, transaction flows, and approvals
- Satellites historize changes in account status, credit scores, and fraud flags
- Enables KYC/AML compliance, audit trails, and credit risk modeling


4. **Retail**

A global retailer integrates point-of-sale, inventory, and online behavior data:

- Hubs represent customers, products, stores, and campaigns
- Links relate purchases, promotions, and customer interactions
- Satellites track inventory levels, pricing changes, and loyalty metrics
- Supports omni-channel analytics and dynamic pricing strategies


5. **Manufacturing**

An industrial manufacturer merges sensor data, production logs, and supplier records:

- Hubs for machines, parts, production orders, and suppliers
- Links show component usage, production chains, and supplier dependencies
- Satellites record machine conditions, part specs, and quality checks
- Business Vault enables predictive maintenance and supply chain resilience


**Conclusion:**

Data Vault 2.0 is not just a model; it is an enterprise-ready way of working that supports long-term data strategy, enables rapid delivery, and enforces trust in analytics. It equips data teams with a framework that is as resilient as it is scalable.