Business Agility Requires Multiple Approaches
Agile businesses create business agility through a combination of business decision agility, time-to-solution agility and resource agility.

This article addresses how data virtualization delivers time-to-solution agility. Part 1 addressed business decision agility and Part 3 will address resource agility.

Time-To-Solution Agility = Business Value
When responding to new information needs, rapid time-to-solution is critically important and often results in significant bottom-line benefits.

Proven, time and again across multiple industries, substantial time-to-solution improvements can be seen in the ten case studies described in the recently published Data Virtualization: Going Beyond Traditional Data Integration to Achieve Business Agility.

Consider This Example: If the business wants to enter a new market, it must first financially justify the investment, including any new IT requirements. Thus, only the highest ROI projects are approved and funded. Once the effort is approved, accelerating delivery of the IT solution also accelerates realization of the business benefits and ROI.

Therefore, if incremental revenues from the new market are $2 million per month, then the business will gain an additional $2 million for every month IT can save in time needed to deliver the solution.

Streamlined Approach to Data Integration
Data virtualization is significantly more agile and responsive than traditional data consolidation and ETL-based integration approaches because it uses a highly streamlined architecture and development process to build and deploy data integration solutions.

This approach greatly reduces complexity and reduces or eliminates the need for data replication and data movement. As numerous data virtualization case studies demonstrate, this elegance of design and architecture makes it far easier and faster to develop and deploy data integration solutions using a data virtualization platform. The ultimate result is faster realization of business benefits.

To better understand the difference, let's contrast these methods. In both the traditional data warehouse/ETL approach and data virtualization, understanding the information requirements and reporting schema is the common first step.

Traditional Data Integration Has Many Moving Parts
Using the traditional approach IT then models and implements the data warehouse schema. ETL development follows to create the links between the sources and the warehouse. Finally the ETL scripts are run to populate the warehouse. The metadata, data models/schemas and development tools used within each activity are unique to each activity.

This diverse environment of different metadata, data models/schemas and development tools is not only complex but also results in the need to coordinate and synchronize efforts and objects across them.

Experienced BI and data integration users will readily acknowledge the long development times that result from this complexity, including Forrester Research in its 2011 report Data Virtualization Reaches Critical Mass.

"Extract, transform, and load (ETL) approaches require one or more copies of data staged along the physical integration process flow. Creating, storing, and manipulating these copies can be complex and error prone."

Data Virtualization Has Fewer Moving Parts
Data virtualization uses a more streamlined architecture that simplifies development. Once the information requirements and reporting schema are understood, the next step is to develop the objects (views and data services) used to both model and query the required data.

These virtual equivalents of the warehouse schema and ETL routines and scripts are created within a single view or data service object using a unified data virtualization development environment. This approach leverages the same metadata, data models/schemas and tools.

Not only is it easier to build the data integration layer using data virtualization, but there are also fewer "moving parts," which reduces the need for coordination and synchronization activities. With data virtualization, there is no need to physically migrate data from the sources to a warehouse. The only data that is moved is the data delivered directly from the source to the consumer on-demand. These result sets persist in the data virtualization server's memory for only a short interval.

Avoiding data warehouse loads, reloads and updates further simplifies and streamlines solution deployment and thereby improves time-to-solution agility.

Iterative Development Process Is Better for Business Users
Another way data virtualization improves time-to-solution agility is through support for a fast, iterative development approach. Here, business users and IT collaborate to quickly define the initial solution requirements followed by an iterative "develop, get feedback and refine" process until the solution meets the user need.

Most users prefer this type of development process. Because building views of existing data is simple and fast, IT can provide business users with prospective versions of new data sets in just a few hours. The user doesn't have to wait months for results while IT develops detailed solution requirements. Then business users can react to these data sets and refine their requirements based on the tangible insights. IT can then change the views and show the refined data sets to the business users.

This iterative development approach enables the business and IT to hone in on and deliver the needed information much faster than traditional integration methods.

Even in cases where a data warehouse solution is mandated by specific analytic needs, data virtualization can be used to support rapid prototyping of the solution. The initial solution is built using data virtualization's iterative development approach, with migration to the data warehouse approach once the business is fully satisfied with the information delivered.

In contrast, developing a new information solution using traditional data integration architecture is inherently more complex. Typically, business users must fully and accurately specify their information requirements prior to any development, with little change tolerated. Not only does the development process take longer, but there is a real risk that the resulting solution will not be what the users actually need and want.

Data virtualization offers significant value, and the opportunity to reduce risk and cost, by enabling IT to quickly deliver iterative results that enable users to truly understand what their real information needs are and get a solution that meets those needs.

Ease of Data Virtualization Change Keeps Pace with Business Change
The third way data virtualization improves time-to-solution agility is ease of change. Information needs evolve. So do the associated source systems and consuming applications. Data virtualization allows a more loosely coupled architecture between sources, consumers and the data virtualization objects and middleware that integrate them.

This level of independence makes it significantly easier to extend and adapt existing data virtualization solutions as business requirements or associated source and consumer system implementations change. In fact, changing an existing view, adding a new source or migrating from one source to another is often completed in hours or days, versus weeks or months in the traditional approach.

Conclusion
Data virtualization reduces complexity, data replication and data movement. Business users and IT collaborate to quickly define the initial solution requirements followed by an iterative "develop, get feedback and refine" delivery process. Further independent layers make it significantly easier to extend and adapt existing data virtualization solutions as business requirements or associated source and consumer system implementations change.

These time-to-solution accelerators, as numerous data virtualization case studies demonstrate, make it far easier and faster to develop and deploy data integration solutions using a data virtualization platform than other approaches. The result is faster realization of business benefits.

Editor's Note: Robert Eve is the co-author, along with Judith R. Davis, of Data Virtualization: Going Beyond Traditional Data Integration to Achieve Business Agility, the first book published on the topic of data virtualization. This series of three articles on How Data Virtualization Delivers Business Agility includes excerpts from the book.

• How Data Virtualization Improves Business Agility – Part 3
• Data Virtualization at Pfizer: A Case Study
• Data Virtualization at Northern Trust: A Case Study
• How Data Virtualization Delivers Business Agility – Part 1