data_services

Data Services Layer White Paper: Data Services Architecture

Build the Enterprise Data Foundation for Your SOA

Nearly every technology analyst firm includes some form of a data services layer in their SOA roadmap. Gartner and AMR Research call this critical capability a data services layer. Forrester calls it an information fabric or data virtualization. IDC calls it an information services layer. The Linthicum Group calls it a data abstraction layer. All agree it is important.

"No SOA plan is complete without a data services layer."

Source: Successful Service-Oriented Architectures Build in a Data Services Layer
November 10, 2005
Eric Austvold, Jeff Hojlo
Copyright © 2005 AMR Research, Inc.

Figure: Data Services Layer

Data Services Layer In Action
A data service layer is especially valuable when there is a large demand for new applications that leverage existing diverse data.

Financial Research Data Services Layer. Data services form a middle layer between a multi-terabyte financial research database and a variety of Matlabs analytical and custom financial engineering applications. Analysts spend less time on access and more time on analysis resulting in higher investment returns.
Reporting Data Services Layer. Multiple reporting requirements including Prime Brokerage, Reconciliation, Risk Management, etc., share a common set of data services to accelerate new reporting development while improving reporting accuracy.
Single View of a “Person of Interest”. Intelligence data spread across multiple intelligence agencies leverage a shared data services layer enabling analysts to get a complete picture on persons of interest.

Data Service Layer Delivers Real Benefits
According to a recent IDC study, enterprises can achieve significant business and technical benefits when they adopt a standardized data services layer in conjunction with their SOA strategy.

Business Benefits

Promote enterprisewide information consistency, visibility, and governance
Agility to address new business rules and requirements with a configurable environment
Increase accessibility of information for critical stakeholders
Support for real- or right-time operational or analytical information processing
Ability to quickly leverage existing information resources in new ways
Decrease time to market with capabilities to support new or enhanced business needs

Technical Benefits

Invoke and manage reusable, standardized, individual, or composite information services from a centralized location for multiple purposes
Optimize performance and information-centric processing activities; reduce round-trips to the database, client CPU, and memory
Protect source data from inappropriate access, erroneous use, and overload conditions
Apply consistent sets of information policies and procedures
Separate applications from the detailed complexities and technical complexities involved in data access, update, and integrity processes
Insulate data and content source changes; perform maintenance and alterations to information sources without disrupting consuming applications

Run Time Operations are the Key to Success
Data services have both a build and run aspect. Once you have built your data services, the key challenge is efficient and effective operations. This requires a centralized home to manage and run your data services, high performance to support heavy-duty queries, flexible deployment to fit with existing management policies and procedures, and open integration to interoperate with other SOA components such as enterprise service buses (ESBs) and governance applications.

Composite: Best of Breed Data Services Layer
The Composite Information Server, along with the Active Cluster Option provides a complete solution for your data service layer needs.

Run Time Repository and Management System. Manage and reuse shared data services along with shared metadata.
High Performance. Optimize query processing. Leverage caching. Apply clustering to add resources.
Flexible Deployment. Conform to your IT Operations processes and policies Extend from project to enterprisewide SOA.
Open Integration. Interoperate with any Application Server, ESB, Governance, and BPM middleware. No dependencies.