Database Systems: Chapter 2 - Data Models

Front 1

What is Data Modelling?

Back 1

The first step in designing a database; process of creating a specific data model for a determined problem domain.

Front 2

What is a Problem Domain?

Back 2

A clearly defined area within a real-world environment with well defined cope/boundaries. To be systematically addressed.

Front 3

What is a Data Model?

Back 3

A relatively simple representation, usually graphical of more complex real-world data structures; a data model represents data structures and their characteristic relations, constraints, transformations and other constructs.

Front 4

What is a Model?

Back 4

An abstraction of a more complex real-world object or event whose main function is to help one understand the complexity of the real-world environment.

Front 5

What should a final/implementation ready Data Model contain?

Back 5

a description of the data structures that will store end-user data; a set of enforceable rules to guarantee the integrity of the data; a data manipulation methodology to support the real-world data transformations.

Front 6

What is the importance of Data Models?

Back 6

They facilitate interaction among designer, applications programmer and the end-user; foster improved understanding of the organisation in question; a communication tool; provide a whole of system view/overall database blueprint; one cannot build a good database without having a appropriate data model.

Front 7

What are the basic building blocks of the Data Model?

Back 7

entities: anything about which data are to be collected; attributes: characteristics of an entity e.g. name, phone, dob (equivalent of fields in the file system); relationships: describe an association among entities (1:1, 1:M, M:N); constraints: restrictions placed on data to help ensure data integrity, often in the forms of rules (e.g. salary between x & y).

Front 8

What are Business Rules?

Back 8

Brief, precise & unambiguous description of a policy, procedure or principle within a specific organisation; derived from a detailed description of an organisations operations, they help to create and enforce actions within that environment; used to define entities, relationships and constraints; describe in simple language the main and distinguishing characteristics of the data as view by the organisation.

Front 9

What are the main sources of Business Rules?

Back 9

Company managers, policy makers, Dept managers and written documentation(procedures, standards and operational manuals); faster and more direct is interviews with end-users, although often less reliable to to differing perspectives.

Front 10

Why is indentifying and documenting Business Rules important?

Back 10

Help standardise an organisation's view of data; can be a communication tool between users and designers; allows designer to understand the nature, role and scope of the data; allow the designer to understand the business processes; allow designer to develop appropriate relationship participation rules and constraints and to create an accurate data model.

Front 11

How do you translate Business Rules into Data Model components?

Back 11

a noun in a business rule will translate to an entity; a verb will translate to a relationship; e.g " a company generates many invoices"; to properly identify relationship type, ask " How many instances of B are related to one instance of A?" and " How many instances of A are related to one instance of B?"

Front 12

List the major Data Models.

Back 12

File system (60s-70s) ; hierarchical and network data models (70s); relational data model (70s - now); object -oriented and extended relational data models (80s-now); XML.

Front 13

Summarise the Hierarchical Data Model.

Back 13

Developed in 60s to manage large amounts of data for complex manufacturing projects; basic logical structure represented by an upside-down tree; contains levels or segments - equivalent of a file system record type (parent and child segments); model depicts a set of 1:M between parents and its children.

Front 14

What are the pros and cons of the Hierarchical Data Model?

Back 14

Pros: big improvement over file system model; many features formed foundation of current data models; became popular/dominant in 70 with many tried and true applications. Cons: complex to implement, difficult to manage, lacked structural independence. Cannot support data relationships other than 1:M. No standards on how to implement model.

Front 15

Summarise the Network Data Model.

Back 15

Created to represent complex data relationships mode effectively that the hierarchical model, to imporve database performance and impose a database standard. Users percieve the network database as a collection of records in 1:M relationships. Unlike the hierarchical model, it allows a record to have more than one parent. Each relationship is called a set and each set is composed of an owner and a member record.

Front 16

What are the limitations of the Network Data Model?

Back 16

As information needs grew and more sophisticated databases and applications were required, the network model grew cumbersome. Lack of ad hoc query capability put heavy pressure on programmers to generate even the simplest of reports. Although had limited data independence, any structural changes still caused havoc in application programs. Both hierarchical and network data models were largely replaced by relational model in the 80s.

Front 17

What was CODASYL?

Back 17

Conference on Data Systems Languages: created the Database Task Group (DBTG) in the late 60s. The DBTG was charged to define standard spec for an environment that would facilitate database creation and data manipulation. The final DBTG report contained specs for three crucial data base components.

Front 18

What were the DBTG report specifications?

Back 18

The schema - conceptual organisation of the entire database as viewed by the administrator. Includes definitions of database name, record type for each record and the components that make up those records. Subschema - defines portion of the database "seen" by application programs. Allows them to simply invoke the subschema required to access the appropriate database files. Data manipulation Language (DML): defines envoriment in which the data can be manager DBTG specified 3 distinct DML components: schema data definition language (DDL), subschema DDL and a DML for working with data.

Front 19

Summarise the Relational Model.

Back 19

Introduced 1970 by E.F.Codd (of IBM); represented a major breakthrough for users/designers; "automatic transmission" compared to "manual transmission" of predecessor databases; foundations is mathematical concept of the "relation" (table) as a matrix composed of intersecting rows and columns; each row is a "tuple" each column is an attribute; initially impractical due to computing overhead; relations are related through sharing of common entity characteristics (values in columns) / common attributes.

Front 20

Discuss the RDBMS.

Back 20

The relational model is implemented through sophisticated relational database management system; performs same basic functions provided by hierarchical and network DBMS systems in addition to a host of other functions that make the relational data model easier to understand and implement; importantly, ability to hide the complexities of relational model from the user. The RDBMS manages all physical details and user sees database as a collection of tables in which the data are stored; user can manipulate/query data intuitively/logically.

Front 21

What is a Relational Diagram?

Back 21

A representation of the relational database's entities, the attributes within those entities and the relationships between those entities.

Front 22

Discuss the Relational Table.

Back 22

A relational table stores a collection of related entities. Therefore a relational database table resembles a file however the crucial difference -> relational table yields complete data and structural independence due to its purely logical structure.

Front 23

How did SQL contribute to the dominance of the Relational Data Model?

Back 23

SQL (as a 4th gen language) allows users to specify what must be done without specifying how is must be done; RDBMs use SQL to translate user queries into instructions for retrieving the requested data. Any SQL-based RDBMS involves three parts: - end-user interface; collection of tables stored in database; SQL engine

Front 24

What is the Entity Relationship Model (ERM)?

Back 24

A data model first developed by Peter Chen in 1975; describes relationships (1:M. 1:1: M:N) among entities at the conceptual level with the help of entity relationship diagrams (ERDs); easier to examine structures graphically that in text.

Front 25

What is an Entity Relationship Diagram?

Back 25

d Diagram that depicts and entity relationship model's entities, attributes and relationships; graphical representations to model database components; using original Chen notation and more current Crow's foot notation (design standard).

Front 26

What are the components of the ERM?

Back 26

Entities = anything about which data are to be collected an stored; represented in ERD by rectangle or entity box; Usually when applying the ERD to the relational model, an entity is mapped to a relational table. Each row is an entity instance or occurence. Each entity is described by a set of attributes. Relationships = describe associations among data, usually between 2 entities. The term connectivity use to label relationship types (1:M, 1:1, M:N).

Front 27

Discuss the Object-Oriented Data Model.

Back 27

Increasingly complex real-world problems demonstrated a need for a data model that more closely represented the real world. In OODM, both data and relationships contained in a single structure called an object. OODM said to be a semantic data model. Unlike relational entities, objects include information about relationships between facts within objects as well as between objects -> gives greater meaning. Objects include all operations that may be performed on it (self-contained).

Front 28

What are the components of the OODM?

Back 28

Object = abstraction of a real world entity; may be considered to be the equivalent of the ER model's entity. An object represents only one occurrence of an entity; Attributes: define the properties of an object (name, dob etc.); Objects that share similar characteristics are called classes, which are collections of similar objects with shared structures and behaviours (methods). A method represents a real world action and define object's behaviour. Classes are organised in a class hierarchy (upside-down tree in which each class has only one parent). Inheritance is the ability of a class to inherit the attributes and methods of the classes above it.

Front 29

How are OODMs modelled?

Back 29

Object-oriented data models are typically depicted using UML (Unified Modelling Language) class diagrams. UML is a language based on OO concepts that describe a set of diagrams and symbols that can be used to graphically mode a system. UNL class diagrams are used to represent data and their relationships within the larger UML object oriented systems modelling language.

Front 30

What is the Extended Relational Data Model?

Back 30

The ERDM is another semantic data model developed in response to the increasing complexity of applications. It is the relational models respondse to the OODM and includes many of the OODMs best features within in an inherently simpler relational database structure environment. The ERDM is primarly geared to wards business applications and the OODM tends to focus on specialised engineering and scientific applications.

Front 31

What are the common characteristics of a data model in order for it to be widely accepted?

Back 31

must show some degree of conceptual simplicity without compromising the semantic completeness of the model; must represent the real world as closely as possible; representation of the real-world transformation(behaviours) must be in compliance with the consistency and integrity characteristics of any data model.

Front 32

What are the levels of the data abstraction framework?

Back 32

external, conceptual, internal and physical

Front 33

What is the External Model?

Back 33

End user's or application programmer's view of the data environment. Has a business/modular focus and often works with a data subset of the global database schema.

Front 34

What is the conceptual model?

Back 34

- a global view of
an entire database.
- integrates all external views
- describes the main data objects,
avoiding details.

Front 35

What is the internal model?

Back 35

- representation of databsae as seen by the DBMS
- a level of
data abstraction that adapts the conceptual model to a
specific DBMS model for implementation.

Front 36

What is the physical model?

Back 36

- operates at the lowest level of abstraction, describing the way data area saves to physical media