Browse Prior Art Database

Terminology and an Architecture on Data Independence

IP.com Disclosure Number: IPCOM000076941D
Original Publication Date: 1972-May-01
Included in the Prior Art Database: 2005-Feb-24
Document File: 4 page(s) / 57K

Publishing Venue

IBM

Related People

Meltzer, HS: AUTHOR

Abstract

The described architecture involves two independent concepts related to logical data structure and physical data organization to implement dynamic data independence. Data independence is a property of a data system which permits variations in the representation and organization of data as stored without requiring reprogramming. It also permits variations on the program's view of the data without requiring rearranging the data. Static data independence is provided by an implementation which requires that if a change to the stored data need be made, then new data descriptors must be written, all existing data must be modified to conform to the new data descriptors, and all programs which use this data recompiled to conform to the new data descriptors.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 38% of the total text.

Page 1 of 4

Terminology and an Architecture on Data Independence

The described architecture involves two independent concepts related to logical data structure and physical data organization to implement dynamic data independence. Data independence is a property of a data system which permits variations in the representation and organization of data as stored without requiring reprogramming. It also permits variations on the program's view of the data without requiring rearranging the data. Static data independence is provided by an implementation which requires that if a change to the stored data need be made, then new data descriptors must be written, all existing data must be modified to conform to the new data descriptors, and all programs which use this data recompiled to conform to the new data descriptors. Dynamic data independence provided by an implementation in which the system consults the data descriptors as it traverses its file, thus permitting variations along the file. This permits partially rearranged data and permits different performance factors or priorities to be applied to different portions of a file.

The first concept is a set of definitions which clearly identifies the interests of the various parties in a data base environment, and identifies the interfaces. The second concept is an architecture of tables, control blocks and descriptors associated with a quantity of stored data. These tables include, essentially, descriptors of data storage, an index to this data and to the associated descriptors, an intermediate descriptor which is used in both name resolution and mapping algorithm, a program work area descriptor and a mechanism for naming the particular logical record required. These descriptors are unbound and are associable in many-to-many relationship. The mapping of the stored records onto the physical records is a function of a data organization and not this architecture. Thus, this architecture is applicable to providing data independence on any data set organization. The following definitions apply. An entity record is the conceptual collection of all information, for which a representation can be constructed or recorded in a data base about an entity, that is, data which is logically related. A logical record is a collection of one or more fields as viewed by an application program. By definition, all of the fields in a logical record represent information about the same entity. Thus, it is a subset of an entity record. Each entity and thus each logical record has one unique identifier. A collection of logical records comprise a file. A stored record is a collection of related data elements upon a medium. These relationships depend on the storage organization, and may be independent of the logical relationships of the various entity records about which information is stored. A collection of stored records comprises a data set. A physical record is a unit of recording upon a medium such as between gaps on ta...