Entity Relationship Diagrams. Supplier Shipment Item Production Plan Product Master Schedule Sent by Sends Supplies Supplied by Produced on Produces Generated.
Post on 20-Dec-2015
217 Views
Preview:
Transcript
Supplier
Shipment Item ProductionPlan
ProductMaster
Schedule
Sent by
Sends
Supplies
Supplied by
Produced on
Produces
Generated on
Generates
Builds
Built on
Composed of
Goes into
Included on
Includes
Mandatory 1
Mandatory many
Optional many
Many with maximum
Entity Relationship Diagrams
• Purpose: Show the structure (relationship of data elements)
• Can be used in multiple stages of development
• Used to communicate and verify understanding between developers and users, and to document the perceived data model
ERD Elements• Entities
– Things about which you collect information
• Relationships– Means of association between entities
• Identifiers– Unique attributes of the entity
• Attributes– Characteristic or property of the entity that is of
interest
ERD Symbols
• Entities: rectangle
• Relationships: diamond (may or may not be used), on a line showing the “cardinality” of the relationship (1 to many), etc.
• Identifiers: Underlined text
• Attributes: Text by the entity (if shown at all on the diagram)
Sample ERDs
1
1
M
1
ORDER
CAN HAVE
PART
SUPPLIER
CAN HAVE
Entity attributes:
ORDER: #, DATE, PART #, QUANTITY
PART: #, DESCRIPTION, UNIT PRICE, SUPPLIER #
SUPPLIER: #, NAME, ADDRESS
Order
Part
Supplier
ERD Notes
• Relationship connect entities
• Attributes are not technically a part of the diagram but are often included
• ERD is not the same as a relational table design
Entity Sets
• Each individual object is called an entity. A collection of such entities is an entity set.– Example: Joe, Jill, and Mary represent entities.
They are all ascribed to the entity set, PERSON.
– Example: A collection of projects is the project entity set.
Relationship
• Relationship– One interaction between one or more entities– For example: if a person works on a project,
there is a relationship between that person and the project
• Relationship set– A collection of such relationships.– A component in an E-R diagram that represents
a set of relationships with the same properties.
Notes on Entities and Relationships
• We can actually see entities, but we cannot see relationships.– For example:
• Entity sets: Person; project
• Relationship: People work on projects. This becomes the relationship set “Work” Persons
Projects
WorkThe set of people, set of projects and set of working relationships.
Entity-Relationship Structures
Persons
Warehouses
SuppliersParts
ProjectsDepts
Are-In Work-On Use Hold
Supply
More complex relationship:See persons and parts
Projects use partsSuppliers supply partsWarehouses hold parts
Persons are in departmentsPersons work on projects
More than two entity sets can be associated with the same relationship set
Customers
Parts
StoresBuy
Each relationship in this set includes a person, a part boughtby the person, and the store where the purchase was made.
Relationship sets that include only twoentities are known as binary. More thantwo are known as N-ary.
A person, a part bought by the person,and the store where the purchase was made
Identifiers
• One of the attributes of an entity or relationship set is called the identifier
• It has one important property: its values identify unique entities in the entity set.
Identifiers are underlined here
Persons
Projects
Work
The set of people, set of projects and set of working relationships.
PERSON-IDNAMEADDRESS
PERSON-IDPROJECT-IDTIME-SPENT
PROJECT-IDSTART-DATEBUDGET
Convention for Identifiers in Relationships
• Use the identifiers of the entities that participate in the relationship as the relationship identifiers.
• Identifiers are not file keys here. At this stage, they are the identifiers of entities that participate in the relationship.
Cardinality
• The number of relationships in which one entity can appear.
• An entity can appear in:– one (1) relationship;– any variable number (N) of relationships; and– a maximum number of relationships
Cardinality - Example
Persons
Projects
Work
The set of people, set of projects and set of working relationships.
PERSON-IDNAMEADDRESS
PERSON-IDPROJECT-IDTIME-SPENT
PROJECT-IDSTART-DATEBUDGET
N
M
A persona can appear in more than oneWORK relationship, and so can a project.If there was a limit to the number of times anentity can take part in the relationship, thenN or M would be replaced by the actualmaximum number.
Cardinality - Example
Manager
Projects
Manage
The set of people, set of projects and set of working relationships.
MANAGER-IDNAMEADDRESS
PERSON-IDPROJECT-IDTIME-SPENT
PROJECT-IDSTART-DATEBUDGET
1
N
Here a project has one (1) manager, whereasa manager can manage any number (N) of projects.
Cardinality - Example
Manager
Projects
Manage
The set of people, set of projects and set of working relationships.
MANAGER-IDNAMEADDRESS
PERSON-IDPROJECT-IDTIME-SPENT
PROJECT-IDSTART-DATEBUDGET
1
N
The denotes optional participation on theproject. If it is mandatory, then there is no placed there.
Issues in building an ER-Diagram
• How to choose entities, relationships and attributes
• How to choose names
• What steps should be followed
Choosing Attributes
• Attributes, just like entity and relationship sets, should express simple concepts.
• E-R diagrams should not contain multi-valued or structured attributes– For example:
PERSONSPERSON-IDDATE-OF-BIRTHQULIFICATION* (asterisk means it is multivalued-repeating)ADDRESS(NUMBER, STREET, SUBURB)Non-simple attributes
Addresses has structured attributes.These should be replaced in the final diagramby relationships.
Ex: Removing multi-valued and structured attributes
Persons
ADDRESSESQUALIFICATIONS
HAVE LIVE-AT
PERSON-IDDATE-OF-BIRTH
PERSON-IDNUMBERSTREETSUBURB
NUMBERSTREETSUBURB
PERSON-IDQUALIFICATION
QUALIFICATION
Choosing Object Set Names
• Remember, that one goal of E-R modeling is to produce a model that is easily understood by users as well as computer personnel.– Entity sets are labeled as nouns– Relationship sets are labeled by verbs– Relationship sets are structured as prepositions
when modeling structural relationships
Where to begin
• Start with entity sets
• Look at how entities interact with each other and model this in terms of relationship sets
• Then add cardinality to the system
• Add attributes and choose identifiers
Normalization
• The process of organizing data in a database. This includes creating tables and establishing relationships between those tables according to rules designed both to protect the data and to make the database more flexible by eliminating two factors: redundancy and inconsistent dependency.
Factor 1: Inconsistent Dependency
• Customer table should have customer address but not employee salary who calls on customer
• If customer information is contained in an order file, if the order is canceled, all the customer information could be lost– Solution: create two tables--one table contains
order information and the other table contains customer information.
Factor 2: Redundant Data
• Data exists in more than one place
• Wastes disk space and creates maintenance problems– Example: a staff person changes their telephone
number and every potential customer that person ever worked with has to have the corrected number inserted.
Rules for Database Normalization
• First Normal Form – No repeating groups
• Second Normal Form – Eliminate data redundancy
• Third Normal Form: – Eliminate data not dependent on key
Database Normalization - Example
Employee NameEmployee IDDepartmentDept AddressItem#Item DescriptionItem PriceWarehouse IDWarehouse AddressItem Location in each WarehouseQuantity on Hand in each Warehouse
ISDATADDMOD
STARTING WITH A SET OF DATA ITEMS:
Database Normalization - Example
Employee ID
Employee Name
Department
Dept Address
Item#Item DescriptionItem PriceWarehouse IDWarehouse AddressItem Location in each WarehouseQuantity on Hand in each Warehouse
ISDATADDMOD
1. CLUSTER DATA ITEMS INTO ENTITIES (to become TABLES):
Database Normalization - Example
Item# Description Price WarehouseID Address ItemLocation Quantity on Hand WarehouseID Address ItemLocation Quantity on Hand WarehouseID Address ItemLocation Quantity on Hand WarehouseID Address ItemLocation Quantity on Hand
Item# Description Price
Item# Warehouse ID Address ItemLocation Quantity on Hand
From:
To:
NOTE: Item# propagates down and becomes part of the identifier. Why?
ISDATADDMOD
2. PULL OUT MULTI-VALUED ITEMS or REPEATING GROUPS:
Database Normalization - Example
Item# Description PriceItem# Warehouse ID Item Location Quantity on Hand
From:
To:
Item# Description Price
Item# Warehouse ID Address Item Location Quantity on Hand
Warehouse ID Address
ISDATADDMOD
3. PULL OUT FACTS ABOUT A PORTION OF THE KEY (partial dependency):
Database Normalization - Example
From:
To:Employee ID Employee Name Department Dept
Address
Employee ID Employee Name Department
Department ID Dept Address
What is the “Department” field called in the Employee record?
Why does it remain in the Employee record?
ISDATADDMOD
4. PULL OUT FACTS ABOUT A NON-KEY DATA ITEM (transitive dependency):
top related