6.3.3 data models

• The entity relationship model does what?

  Represents relationships between different entities in a system
• Entity relationship models use entities (that represent real world concepts etc), attributes, and relationships
• Entity relationship models are useful to visualize and design the structure of a database by illustrating relations between them
• Data models ensure that a database design will be fit for purpose and without omission
• In ERD's how are entities, represented?

  Tables
• In ERD's how are relationships represented?

  By arrows between entities
• What are data dictionaries?

  Tables that contains formatting data about the data (metadata)
• Data dictionaries are used to describe the data that will be held in a data model
• In ERD's how are attributes represented?

  By the items within the tables
• What are the three levels that data modelling occurs at?

  Physical

  Conceptual

  Logical
• What is the physical level of data modelling:

  Framework for how data is stored in a database
• What is the conceptual level of data modelling:

  The high-level user view of data
• What is the logical layer of data modelling:

  The logical layer of data modelling sits between conceptual and physical, representing data logically separate from it's physical components
• One to one relationships occur when a business needs to use entities that are super and sub types
• What are subtypes: Types sourced from another type
• An example of supertype and subtype is:

  ‘product’ as supertype and ‘book’ as subtype.​
• One-to-many relationships occur when one thing has many subordinate parts.
• An example of one-to-many relationship is:

  If vendor A makes two products then both products are related to the vendor but neither are related to each other
• Many-to-many relationships occur between things with multiple relations directly to each other.
• An example of a many-to-many relationship is between an author and their books, this is because the author is directly connected to each of their books; additionally, however, they may have co-authored in which case the book has two direct connections
• To break up many-to-many entities (as it is very difficult to code for them), what method is used?

  Adding an additional entity that represents the overlap
• A solution to the example of books and authors, would be to create the entity author-book, which will hold the id of the book entities and the corresponding author entities
• The advantages of entity relationship models are that:

  It is easy to interpret

  Easy to simulate the real relationships

  Easy to create a ER model software, making it easier to edit, collaborate over, and
• The disadvantages of entity relationship models are that:

  There is no industry standard

  Doesn't represent data manipulation

  Inefficient representation of many to many relationships
• UML is short for unified modelling language, but what does it model?

  Many things system and software developers need
• What are the advantages of UML:

  Allows many groups to collaborate as they all operate on a consistent language

  Easy to use to model as it as you don't have to design your own language

  Independent of particular programming languages and development processes.
• What are the disadvantages of UML:

  Because it is so consistent you cannot edit to suit your needs, without removing it's compatibility

  Unnecessarily detailed for some diagrams

  Must first be understood before being used
• UML is great for modelling object-orientated programming
• UML is very useful for many things, which means that sometimes it may feel bloated
• Data dictionaries advantages:

  Provides clear, concise, and structured info regarding databases

  Once a data dictionary of a database is understood it is far easier to use the database

  Databases are often very large, and complex, data dictionaries abstract the unnecessary detail
• Data dictionary disadvantages:

  Data dictionaries are difficult for the technologically illiterate to understand

  Creating a new data dictionary is a very big task. It may take years to create one.

  Data dictionaries are expensive to make and maintain
• What kind of diagram is this?

  Physical

  
• How are relationships between entities defined?

  shared columns, or keys, in the tables.
• What does each entity in a hierarchical data model represent?

  Record like a row of a table
• A storage area network (SAN) or storage network is a computer network which provides access to consolidated, block-level data storage.
• What are super types?

  Type that is the source one or more entities