PD.Relational Data Management: Difference between revisions

From OIAr
Jump to navigation Jump to search
No edit summary
No edit summary
Line 67: Line 67:
}}
}}
{{Table Ending}}
{{Table Ending}}
{{Text Footer GP}}

Revision as of 21:39, 26 November 2019


This is a Pattern Definition document PD Relational Data Management Version: 0.3 OIAr logo
Document type: Pattern Definition Owner:

S.A.D. Jumelet



Description

This Pattern Definition serves the creation of models in the Service Category "Data Management". This Pattern provides the function layout to model "strictly structured data" handling, i.e. it controls the creation, updating and querying of sets of organized data. It is called 'relational data management', because a schema is used to store data in a predefined way, with predefined relations between data objects. The provisioning of the following elements is important with this kind of data handling:

  • a "modeling language", to define schemas;
  • a "query language", to enable data object reading, creating, updating, and deleting;
  • a "transaction mechanism", to enable interaction with other (instances of) data management facilities and applications.

These elements are provided by the Data Engine. Data objects are stored using a predefined (structured) Data Store. Both functions form the core of this Pattern.

Services realized

This Pattern realizes the following service(s):

  • Application Database (This service provides a data store for applications with a predefined structure)

Functional and Integration view

This is the graphic representation of the functional model of this Generic Pattern:

Relational Data Management pattern
Relational Data Management pattern


Pattern Definition Composition

This pattern is an aggregation of the following (mandatory and optional) functions, expressed in Data/Object Types and Fundtion Definitions:

Function Inclusion Description/Rationale
no link Engine recommended The Data Engine is the heart of the pattern, as it offers the intelligence to control strictly structured data handling.
no link Store recommended The Table Store offers the predefined storage structure that is one of the key characteristics of relational data management.
no link Encryption optional When data is exchanged over unsecured data transport facilities, it might be a good measure to encrypt the data at engine level.

Services connected with this Generic Pattern

This Generic Pattern has the following mandatory and optional relations with adjacent Generic Services.

Service Adjacency Summary Rationale
File Storage recommended This service offers clients the ability to store, retrieve and modify data in loosely structured form. Every Data Management pattern must store its data at a lower data level, which means it requires the use of a File Storage service. Even in commercial products that request access to Raw Storage this is true, as these products actually provide the File Storage functionality themselves.
Authentication & Authorization optional This service can validate an identity claim, and it can validate the permissions required for an action, as part of an Authentication & Authorization process. A suitable source for identity and permission information may be required for
  • controlling access to the data stored in the Data Management facility;
  • management of the data;
  • managing the behaviour of the Data Management facility itself, using the Controlling functionality.
Data Protection Management optional This service provides the ability to protect data against accidental or wilful corruption. The data being handled by this Pattern may need to be protected against data corruption and/or loss. Even though this might be delegated to the lower level of the File Storage service that the Pattern uses, it's usually much better to handle data protection at the logical level of this Pattern, as at this level the interdependencies of parts of the datasets is known and controllable.
Archive Management optional This service provides the ability to create and use archives. The Pattern may need to archive (parts of) a structured data set, and/or may require access to an archive containing (parts of) a structured data set, in order to fulfil requests from its clients; furthermore, it may be necessary to migrate (parts of) an active structured data set to an archive. For any of this, the Data Management pattern needs access to an Archive Management service.