What is Data Virtualization?
Data Virtualization is an approach to unify data from multiple sources in a single layer so that applications, reporting tools, and end users can access the data without requiring details about the original source, location, and data structures.
The unified data layer is virtual. Unlike Extract Transform Load (ETL) processes, the majority of the data remains in place, and is queried on-demand directly from the source systems. This reduces mass data movement and brings the end user closer to the original data sources – minimizing the challenges and costs of multi-level data movement and transformation. Data Virtualization is faster and more flexible than traditional Data Integration (DI) – no mass data replication, no new data architecture, no mass data storage.
The unified data layer serves as an abstraction layer for creating data mashups across sources and formats. It reads and analyzes data from multiple sources including structured, unstructured, web, and big data; and makes the data available for integrated querying, reporting, and analytics. Data consumers don’t need a deep understanding of the complex structures and different formats of the underlying data sources. The mashed-up data is accessible through different interfaces including SQL, web services, data dumps/CSV, and pre-formatted reports.
- Enables timely insights across disparate data sources
- Delivers high-performance
- Reduces data replication and storage
- Increases speed and agility of integrated data access
- Reduces development and support time
- Hides complexity of underlying data structures
- Enables multi-channel and multi-mode data access
Data virtualization use cases
The biggest value proposition of data virtualization is quick and efficient access to integrated data from multiple sources. Business users can then generate valuable insights by combining data across sources without having to embark on a major Data Integration project.
Data virtualization is very well suited for:
- Agile Business Intelligence (BI)
- Operational reporting, data visualization, and analytics
- 360 views (customer, product, entity)
- Real-time and point-in-time reporting and analytics
- Integrated data services and Service Oriented Architecture (SOA)
- Data discovery and concept validation (e.g., prior to major DI investment)
Examples of use cases:
- Hospitality – Customer 360 understanding and experience management
- Call Center Operations – Value added customer data and decision support
- Multiple Industry Verticals – Data bridging across data warehouses and data lakes
- Finance/Banking – Real-time fraud detection and warning
- Health Care – Patient 360 views and data services across health information portals
- Real Estate – Automated multi-site metrics and performance measures