Data Modeling pptx. 2

What is data modeling?

The process of creating a specific data model for a determined problem domain.

What is a data model?

A relatively simple representation of more complex real-world data structures.

Why is data modeling an iterative, progressive process?

The model is developed and refined as the database problem and requirements become better understood.

Why are data models important?

They act as a communication tool and provide the overall database blueprint needed to create a sound database.

Why can different users need different views of the same data?

Programmers, managers, and end users may view the same data differently; a good database blueprint provides a common structure.

What is a business rule?

A brief, precise, and unambiguous description of a policy, procedure, or principle within a specific organization.

What do business rules help define?

Entities, attributes, relationships, and constraints.

What are the main sources of business rules?

Company managers, policymakers, department managers, and written documentation such as company procedures.

Why are business rules important to database design?

They standardize the organization's view of data, improve communication between users and designers, clarify the nature/role/scope of data and business processes, and help create relationship rules, constraints, and an accurate data model.

How do you translate a business rule into data-model components?

Identify nouns as objects/entities, identify the verb as the relationship, and use words such as one or many to determine cardinality.

Example: What does 'A customer may generate many invoices' tell you?

CUSTOMER and INVOICE are entities; GENERATE is the relationship; one customer can be associated with many invoices.

What relationship results from 'A customer may generate many invoices' plus 'An invoice is generated by only one customer'?

One-to-many (1:M): one customer can generate many invoices, while each invoice is generated by one customer.

What is data abstraction?

Representing the database at different levels, from the users' view down to physical storage details.

What organization created the data-abstraction framework discussed in the PPT?

The American National Standards Institute (ANSI) Standards Planning and Requirements Committee (SPARC).

What are the four data-abstraction levels in the PPT?

External (user), conceptual, internal (logical), and physical.

What is the order of the four abstraction levels from higher to lower abstraction?

External → Conceptual → Internal (Logical) → Physical.

What is the external model?

The end users' view of the data environment.

What is an external schema?

A specific representation of an external view.

Why do business units have external views?

Applications often have a specific business-unit focus, so users see subsets of the database relevant to their unit.

What represents external views?

ER diagrams can be used to represent external views.

What are the advantages of external views?

They identify data needed by each business unit, provide feedback about model adequacy, help ensure security constraints, and simplify application development.

What is the conceptual model?

A global view of the entire database by the entire organization.

What is another name for the conceptual model?

Conceptual schema.

What is the conceptual model used for?

Identifying and providing a high-level description of the main data objects.

What is the most widely used conceptual model?

The Entity Relationship (ER) model.

What are the advantages of the conceptual model?

It gives an easy-to-understand bird's-eye view of the data environment and is independent of both software and hardware.

What is the internal (logical) model?

The representation of the database as seen by the DBMS.

What does the designer do when creating the internal model?

Match the conceptual model's characteristics and constraints to those of the selected implementation model.

What is an internal schema?

A specific representation of the internal model using database constructs supported by the chosen database.

Why is the internal model software dependent?

Because it depends on the specific database software being used.

Is the internal model hardware dependent?

No. The internal model is software dependent but hardware independent.

What is logical independence?

The ability to change the internal model without affecting the conceptual model.

What is the physical model?

The lowest level of abstraction; it describes how data is saved on storage media.

What must be defined in a physical model?

Physical storage devices and the physical access methods required to reach data within those devices.

Is the physical model hardware/software dependent?

Yes. The physical model is both hardware dependent and software dependent.

What is physical independence?

The ability to change the physical model without affecting the internal model.

What is a constraint?

A restriction placed on data.

Why are constraints used?

They help ensure data integrity.

Give examples of constraints from the PPT.

Employee salary must be between $6,000 and $350,000; student GPA must be between 0.00 and 4.00; each class must have one and only one teacher.

What are the basic ER/data-model building blocks emphasized in the PPT?

Entities, attributes, relationships, and constraints.

Why was the Entity Relationship Model (ERM) developed?

Complex database design requires conceptual simplicity, and designers need a graphical way to picture entities and their relationships.

What is an Entity Relationship Diagram (ERD)?

A graphical representation used to model database components.

What is an entity?

A person, place, thing, concept, or event about which data will be collected and stored.

How is an entity represented in an ERD?

By a rectangle, also called an entity box.

What is an attribute?

A characteristic of an entity.

What is a relationship?

An association among two or more entities.

What three relationship types are in the PPT?

One-to-one (1:1 or 1..1), one-to-many (1:M or 1..*), and many-to-many (M:N or *..*).

What is a one-to-many (1:M) relationship?

One instance of an entity can be associated with many instances of another entity, while the other side is limited to one.

What is a many-to-many (M:N) relationship?

Many instances on one side can be associated with many instances on the other side.

What is a one-to-one (1:1) relationship?

One instance of one entity is associated with one instance of another entity.

What is cardinality?

The number of instances of one entity that can be associated with instances of another entity.

What is modality?

Whether participation in a relationship is required or optional.

What is the difference between cardinality and modality?

Cardinality asks 'How many?'; modality asks whether participation is required or optional.

What ER notations are identified in the PPT?

Chen notation, Crow's Foot notation, and class diagram notation, which is part of UML.

What does UML stand for?

Unified Modeling Language.

What is the purpose of ER notation?

To graphically represent entities and the relationships among them.

What are good entity-naming conventions?

Entity names should describe objects in the business environment and use terminology familiar to users.

What are good attribute-naming conventions?

Attribute names should clearly describe the data represented.

Why prefix an attribute name with its entity name or abbreviation?

It makes the model easier to understand and helps make it self-documenting; example: CUS_CREDIT_LIMIT for CUSTOMER credit limit.

What do data models represent historically?

Different schools of thought about what a database is, what it should do, what structures it should use, and what technology should implement those structures.

What is the hierarchical model?

A database model developed in the 1960s to manage large amounts of data for complex manufacturing projects.

How is the hierarchical model structured?

As levels or segments arranged in a tree-like parent-child structure.

What is a segment in the hierarchical model?

The equivalent of a file system's record type.

What are parent and child in the hierarchical model?

A higher-level segment is the parent; the segment directly beneath it is the child.

What is the network model?

A model created to represent complex data relationships more effectively than the hierarchical model, improve database performance, and impose a database standard.

What is a schema?

The conceptual organization of the entire database as viewed by the database administrator.

What is a subschema?

The portion of the database seen by application programs that produce desired information from the database.

What is DDL?

Data Definition Language; it enables the database administrator to define schema components.

What is DML?

Data Manipulation Language; it defines the environment in which data can be managed and is used to work with data in the database.

What is the foundation of the relational model?

A mathematical concept called a relation.

What is a relation?

A two-dimensional structure composed of intersecting rows and columns.

In a relational model, what is a row called?

A tuple.

In a relational model, what does a column represent?

An attribute.

What is an RDBMS?

A Relational Database Management System that implements the relational data model.

What does an RDBMS do?

It performs the basic functions provided by hierarchical and network DBMSs and manages the technical details while users see tables containing stored data.

How are relational tables linked?

They can be connected through common attributes/keys; the PPT's example uses AGENT_CODE to connect AGENT and CUSTOMER information.

What is the object-oriented data model (OODM)?

A model in which both data and its relationships are contained in a single structure called an object.

What is the OODBMS?

Object-Oriented Database Management System; the OODM is its basis.

Why is the OODM called a semantic data model?

Because 'semantic' refers to meaning.

What is an object in the OODM?

An abstraction of a real-world entity.

What are attributes in the OODM?

Properties of an object.

What are methods in the OODM?

Operations that can be performed on an object.

What is a class?

A collection of similar objects with shared structure and behavior.

What does a class method represent?

A real-world action, such as finding a person's name, changing a person's name, or printing a person's address.

What is a class hierarchy?

A structure resembling an upside-down tree in which each class has only one parent.

What is inheritance?

The ability of an object within a class hierarchy to inherit the attributes and methods of classes above it.

How are object-oriented data models typically depicted?

Using UML class diagrams.

What is the Internet of Things (IoT)?

A web of Internet-connected devices that exchange and collect data.

How has IoT affected data growth?

It has accelerated the rate of data growth; the PPT states that about 2.5 quintillion bytes of data are created daily.

What is Big Data?

A movement to find new and better ways to manage large amounts of web- and sensor-generated data and derive business insight from it.

What are the 3 Vs of Big Data?

Volume, Velocity, Variety.

What does Volume mean in Big Data?

The amount of data.

What does Velocity mean in Big Data?

The speed at which data is generated/handled.

What does Variety mean in Big Data?

The different types/forms of data.

What is Hadoop?

A Java-based, open-source, high-speed, fault-tolerant distributed storage and computational framework.

What is HDFS?

Hadoop Distributed File System; a highly distributed, fault-tolerant file storage system designed to manage large amounts of data at high speeds.

What is MapReduce?

An open-source application programming interface (API) that provides fast data analytics services.

What is NoSQL?

A large-scale distributed database system that stores structured and unstructured data efficiently.

What are the general characteristics of NoSQL databases?

They are not based on the relational model and SQL; support highly distributed architectures; provide high scalability, high availability, and fault tolerance; support very large amounts of sparse data; and are geared toward performance rather than transaction consistency.

What are the main database models covered in the evolution section?

Hierarchical, network, relational, and object-oriented; the module also covers emerging Big Data and NoSQL approaches.