ADINA (Architecture for Distributed INformation Access) is a research system that enables simplified dynamic access to unrelated data sources through the Internet. The user or the accessing application needs no knowledge of the underlying data source locations, schemas, interrelationships, and query languages.
ADINA is a research system that facilitates seamless access to distinct and unrelated data sources. ADINA lets users and application programmers focus on their tasks—not on the intricacies of the data.
Users and applications request desired data through personalized dynamic (ad hoc) views of the underlying data environment. Users request only the data they want. ADINA determines how to access and navigate the schemas of data sources and how to aggregate the results.
ADINA integrates distinct and heterogeneous information sources, such as relational databases, XML documents, object-oriented databases, image and signal databases, and frame-based knowledge bases. Data sources register with ADINA by describing their schemas and terminology. They are not required to either conform to metadata standards or provide software wrappers.
Provides personalized and real-time configurable views of the integrated data environment
Frees requestors from the need to understand schemas, ontologies, and multiple access languages
Supports ad hoc request capabilities to view information in novel ways
Provides high-level request access that specifies only data to be retrieved and search constraints in terms familiar to the requestor
Access languages are closer to natural languages
Integrates heterogeneous data sources (e.g., relational databases, object-oriented databases, XML documents, frame-based knowledge bases) while preserving their autonomy
Data sources describe themselves to ADINA
Metadata standards not needed
Software wrappers not needed
Provides loose coupling between users and the data
Breaks traditional stovepipes
When data sources change, the application still works!
Scalability-data sources can be easily integrated and modified
What Can It Do?
ADINA provides a powerful environment for virtual data warehousing, data mining, and knowledge discovery. ADINA can be used for numerous and assorted enterprise computing applications, such as E-commerce, finance and accounting, business, science, and manufacturing.
ADINA provides high-level, real-time, and personalized ad hoc access to both new and legacy data sources. ADINA dynamically integrates data sources by managing their interrelationships. Thus, both users and data sources maintain their autonomy because they are insulated from the structures and interrelationships of the data environment. ADINA breaks traditional stovepipes.
Traditionally, application programmers require vast knowledge and skills and must consider many issues when formulating queries to multiple data sources. They must:
Discover which data sources provide the needed information
Understand the underlying schemes and ontologies
Be familiar with multiple access or query languages
Understand the terminologies unique to each data source
Understand how the distributed sources are interrelated
In addition, interface screens are tied to very specific queries, relying on structures of the underlying sources.
ADINA minimizes the application programmer’s burdens by insulating the programmers from this specialized knowledge and understanding of the data environment. Programmers consider only simplified requests and HCI requirements in terminology familiar to them. ADINA handles the specialized data access to multiple sources and aggregates the results.
What Are the Benefits?
Personalized and real-time configurable views of an integrated heterogeneous data environment (relational databases, XML documents, object-oriented databases, frame-based knowledge bases, multimedia, etc.)
Users and applications focus on asking the right questions because ADINA insulates them from underlying data source schemas, ontologies, query languages, interrelationships, and terminology.
Access language is closer to natural language, making the queries easier to express by application programmers.
Data sources remain autonomous because they just “describe” themselves to ADINA. They do not need to conform to metadata standards or provide software wrappers.
Loose coupling between users and data. When data source structures change, the applications still work.
Scalability: diverse sources can be integrated and modified with ease.
How ADINA Works
A user or application formulates a high-level request for data and forwards it to ADINA. ADINA, in turn, automatically formulates and submits complex distributed queries to the pertinent data sources. The results from these data sources are aggregated by ADINA and returned to the user or application.
Traditional Access and ADINA Access to Information Charts