notes of awesome

7/29/2019 Notes of Awesome

1/69

1

High Level Data Modeling with

Entity-Relationship Diagrams

Modeling tools (E/R Diagrams)

Special techniques (Weak Entity Sets)Converting to Relations (Schema Definition)


2/69

2

Purpose of E/R Model

The E/R model allows us to sketchdatabase schema designs.

Includes some constraints, but not operations.

Designs are pictures calledentity-relationship diagrams.

Simple notation - Easy communication

Later: convert E/R designs to relational DBdesigns.

Close to 3NF or BCNF Normal Form


3/69

3

Framework for E/R

Design is a serious business.

Only few times in your life

The boss knows they want adatabase, but they dont know whatthey want in it.

I/O based modeling

Sketching the key components is anefficient way to develop a workingdatabase.

Data model independent design templates


4/69

4

Entity Sets

Entity= thing or object.

Entity set= collection of similar entities.

Similar to a class in object-oriented languages.

Attribute= property of (the entities of) anentity set.

Attributes are simple values, e.g. integers ordecimals, character strings, dates, enumsnot structs, sets, arrays, bags, lists etc.


5/69

5/48

Dataset

2

NoOfWheels

PlateNo

Age

DrivingLicence

Name

Type

Entities

Data components

Entity sets

23

E

Joe

Ferrari 007


6/69

6

Notation methods

ERD: Entity-RelationshipDiagram

Designed for pencil andpaper (Chen, 1970)

UML: Unified ModelingLanguage (Rational, 1990)

Designed for CASE tools

ODL: Object DefinitionLanguage (ODMG, 1997)

Formal language for static

OO design


7/69

7

Example:

Entity set Beers has two attributes, name and

manf(manufacturer).Each Beers entity has values for these two

attributes, e.g. b1: (Bud, Anheuser-Busch)tuple in the relational data model

Beers

name manf

B (finite set)

b 1

b2

b3

b

n


8/69

8

RelationshipsArelationship connects two or more

entity sets.

Binary relationships (ERD, UML, ODL)


9/69

9

Example: Relationships

Drinkers addrname

Beers

manfname

Bars

name

license

addr

Note:license =beer, full,none

Sells Bars sell some

beers.

Likes

Drinkers likesome beers.Frequents

Drinkers frequentsome bars.


10/69

10

Relationship Set

The current value of an entity set is theset of entities that belong to it.

Example: the set of all bars in our database.

The value of a relationship is arelationship set, a set of tuples with one

component for each related entity set.Time is ignored: relationship may occur

between entities any time in the future.


11/69

11

Example: Relationship Set

For the relationship Sells, we mighthave a relationship set like:

Bar BeerJoes Bar BudJoes Bar Miller

Sues Bar BudSues Bar Petes AleSues Bar Bud Lite


12/69

12

Multiway Relationships

Sometimes, we need a relationship thatconnects more than two entity sets.

Only in ERD, but not in UML neither ODL

Suppose that drinkers will only drinkcertain beers at certain bars.

Our three binary relationships Likes, Sells,

and Frequents do not allow us to make thisdistinction.

But a 3-way relationship would.


13/69

13

Example: 3-Way Relationship

Bars Beers

Drinkers

name nameaddr manf

name addr

license

Preferences


14/69

14

A Typical Relationship Set

Bar Drinker BeerJoes Bar Ann Miller

Sues Bar Ann BudSues Bar Ann Petes AleJoes Bar Bob BudJoes Bar Bob Miller

Joes Bar Cal MillerSues Bar Cal Bud Lite


15/69

Decomposition of MultiwayRelationships

15/47

Bars Beers

Drinkers

name nameaddr manf

name addr

license

Preferences


16/69

16

Many-Many (M:N)

RelationshipsFocus: binary relationships,

such as Sells between Bars and

Beers.In a many-many relationship,

an entity of either set can be

connected to many entities ofthe other set.

E.g., a bar sells many beers; a

beer is sold by many bars.

many-many


17/69

17

Many-One (M:1) Relationships

Some binary relationships aremany-onefrom one entity set toanother.

Each entity of the first set isconnected to at most one entity ofthe second set.

But an entity of the second set canbe connected to zero, one, ormany entities of the first set.

many-one


18/69

18

Example: Many-One Relationship

Favorite, from Drinkers to Beers ismany-one.

A drinker has at most one favorite beer.

But a beer can be the favorite of anynumber of drinkers, including zero.

Drinkers BeersFavorite


19/69

19

One-One (1:1) Relationships

In a one-onerelationship, eachentity of either entity set is relatedto at most one entity of the other

set.Example: Relationship Best-seller

between entity sets Manfs

(manufacturer) and Beers.A beer cannot be made by more

than one manufacturer, and nomanufacturer can have more than

one best-seller (assume no ties).

one-one


20/69

Multiplicity Notations

ERD: any kind of theta condition (ie.


21/69

21

Representing Multiplicityin

ERDShow a many-one relationship by an

arrow entering the one side.

Remember: Like a functional dependency.

Show a one-one relationship by arrowsentering both entity sets.

Rounded arrow= exactly one, i.e.,each entity of the first set is related toexactly one entity of the target set.

0..1

1..1


22/69

22

Example: Many-One Relationship

Drinkers BeersLikes

Favorite Notice: two relationshipsconnect the same entitysets, but are different.


23/69

23

Example: One-One Relationship

Consider Best-seller between Manfs and Beers.

Some beers are not the best-seller of anymanufacturer, so a rounded arrow to Manfswould be inappropriate.

But a beer manufacturer has to have a best-seller.

Manfs BeersBest-seller

A manufacturer hasexactly one bestseller.

A beer is the best-seller for 0 or 1manufacturer.


24/69

VisualizingMultiplicity

24/47


25/69

25

Attributes on Relationships

It is allowed to use only in EERDs!

Sometimes it is useful to attach an attributeto a relationship.

Think of this attribute as a property oftuples in the relationship set.

Bars BeersSells

pricePrice is a function of

both the bar and the beer, not of one alone.


26/69

26

Equivalent Diagrams WithoutAttributes on Relationships

Create an entity set representingvalues of the attribute.

Make that entity set participate in the

relationship.Bars BeersSells

price

PricesNote convention: arrowfrom multiway relationship= all other entity setstogether determine aunique one of these.


27/69

27

Equivalent Diagrams WithoutAttributes on Relationships

Create a connecting entity setrepresenting values of the attribute.

Split the relationship into two binaryones.

Bars Beers

price

Prices SellsWhatSellsWhere

A i i d C i Cl


28/69

Association and Connecting Classes

28/47


29/69

29

1-way Relationships

Sometimes an entity set appears morethan once in a relationship.

Label the edgesbetween the

relationship andthe entity set withnames called roles.


30/69

30

Example: Roles

Drinkers

Buddies

1 2

Relationship Set

Buddy1 Buddy2

Bob AnnJoe SueAnn BobJoe Moe

Relationship Set

Husband WifeBob AnnJoe Sue

Drinkers

Married

husband wife

K


31/69

31

Keys

Akey is a set of attributes for one entity

set such that no two entities in this setagree on all the attributes of the key.

It is allowed for two entities to agree on some,

but not all, of the key attributes.We must designate

a key for every entityset.

ERD: exactly one

UML: at most one

ODL: zero or more


32/69

32

Example: a Multi-attribute Key

Courses

dept number hours room

Note that hours and room could also serve as akey, but we must select only one key.


33/69

33

Weak Entity Sets

Occasionally, entities of an entity setneed help to identify them uniquely.

Entity set E is said to be weak if inorder to identify entities ofE uniquely,we need to follow one or more many-

one relationships from E and includethe key of the related entities from theconnected entity sets.


34/69

34

Example: Weak Entity Set

name is almost a key for football players, butthere might be two with the same name.

number is certainly not a key, since playerson two teams could have the same number.

But number, together with the team name

related to the player by Plays-on should beunique.


35/69

35

In E/R Diagrams

Players TeamsPlays-on

name namenumber

Double diamond for supporting many-one relationship. Double rectangle for the weak entity set.

Note: must be roundedbecause each player needsa team to help with the key.


36/69

WeakEntity SetNotations

36

ERD: With supportingrelationships.

UML: Withcompositions.

ODL: There is nosuch thing as

weak class


37/69

37

Weak Entity-Set Rules

A weak entity set has one or moremany-one relationships to other

(supporting) entity sets. Not every many-one relationship from a

weak entity set need be supporting.

But supporting relationships must have arounded arrow (entity at the one end isguaranteed).


38/69

38

Weak Entity-Set Rules (2)

The key for a weak entity set is its ownunderlined attributes and the keys for

the supporting entity sets. E.g., (player) number and (team) name is

a key for Players in the previous example.


39/69

39

Design Techniques

1. Avoid redundancy.

2. Limit the use of weak entity sets.

3. Dont use an entity set when anattribute will do.

4. Use different attribute names.

5. Names are usually case insensitive,but use the camelCase notation style.


40/69

40

Avoiding Redundancy

Redundancy = saying the same thingin two (or more) different ways.

Wastes space and (more importantly)encourages inconsistency.

Two representations of the same fact

become inconsistent if we change one andforget to change the other.

Recall anomalies due to FDs.


41/69

41

Example: Bad

Beers ManfsManfBy

name

This design states the manufacturer of a beertwice: as an attribute and as a related entity.

name

manf

addr


42/69

42

Example: Good

Beers ManfsManfBy

name

This design gives the address of eachmanufacturer exactly once.

name addr


43/69

43

Example: Bad

Beers

name

This design repeats the manufacturers addressonce for each beer and loses the address if thereare temporarily no beers for a manufacturer.

manf manfAddr


44/69

44

Example: Good

Beers ManfsManfBy

name

Manfs deserves to be an entity set because ofthe nonkey attribute addr.Beers deserves to be an entity set because it isthe many of the many-one relationship ManfBy.

name addr


45/69

45

Entity Sets Versus Attributes

An entity set should satisfy at leastone of the following conditions:

It is more than the name of something; ithas at least one nonkey attribute.

or

It is the many in a many-one or many-many relationship.


46/69

46

Example: Bad

Beers ManfsManfBy

name

Since the manufacturer is nothing but a name,and is not at the many end of any relationship,it should not be an entity set.

name


47/69

47

Example: Good

Beers

name

There is no need to make the manufacturer anentity set, because we record nothing aboutmanufacturers besides their name.

manf


48/69

E/R model transformation is an iterative process

1. Identify

entity sets

2. Identify

relationships

3. Determine

multiplicities

4. Determine

attributes

5. E/R diagram

justification

6. Checking,

Validation

T1 Splitting

entity sets

T2 Generalization,

specialization

T3 Division into

independent

entity sets

T4 Using parallel

relationships

T5 Splitting a

relationship

T6 Adding

attributes

T7 Splitting

attributes

vagy


49/69

49

From E/R Diagrams to Relations

Entity set -> relation.

Attributes -> attributes.

Relationships -> relations whoseattributes are only:

The keys of the connected entity sets.

Attributes of the relationship itself.


50/69

50

Entity Set -> Relation

Relation: Beers(name, manf)

Beers

name manf


51/69

51

Relationship -> Relation

Drinkers BeersLikes

Likes(drinker, beer)Favorite

Favorite(drinker, beer)

Married

husband

wife

Married(husband, wife)

name addr name manf

Buddies

1 2

Buddies(name1, name2)

C bi i R l ti


52/69

52

Combining Relations OK to combine into one relation:

1. The relation for an entity-set E2. The relations for many-one relationships

of which E is the many.

Example: Drinkers(name, addr) andFavorite(drinker, beer) combine tomake Drinker1(name, addr, favBeer).

Drinkers BeersFavourite

name addr name manf


53/69

53

Risk with Many-Many Relationships

Combining Drinkers with Likes would bea mistake. It leads to redundancy, as:

name addr beerSally 123 Maple BudSally 123 Maple Miller

Redundancy

Drinkers BeersLikes

name addr name manf


54/69

54

Handling Weak Entity Sets

Relation for a weak entity set mustinclude attributes for its complete key

(including those belonging to otherentity sets), as well as its own, nonkeyattributes.

A supporting relationship is redundantand yields no relation (unless it hasattributes).

E l W k E tit S t


55/69

55

Example: Weak Entity Set ->Relation

Logins HostsAt

name name

Hosts(hostName, location)Logins(loginName, hostName, billTo)

At(loginName, hostName, hostName2)

Must be the same

billTo

At becomes part ofLogins

location


56/69

56

Dont Overuse Weak Entity Sets

Beginning database designers often doubtthat anything could be a key by itself.

They make all entity sets weak, supported by allother entity sets to which they are linked.

In reality, we usually create unique IDs for

entity sets. Examples include social-security numbers,

automobile VINs etc.


57/69

57

When Do We Need Weak

Entity Sets?The usual reason is that there is no

global authority capable of creating

unique IDs.Example: it is unlikely that there could

be an agreement to assign unique

player numbers across all football teamsin the world.


58/69

58

E/Rdiagramof theJoomla 1.5portalsystem

Now let

us go topharmacies


59/69

59

Subclasses

Subclass= special case = fewerentities = more properties.

Example: Ales are a kind of beer. Not every beer is an ale, but some are.

Let us suppose that in addition to all the

properties(attributes and relationships) ofbeers, ales also have the attribute color.


60/69

60

Subclasses in E/R Diagrams

Assume subclasses form a tree.

I.e., no multiple inheritance.

Isa triangles indicate the subclassrelationship.

Point to the superclass.

In an Isa hierarchy, only the root entityset has a key, and it must serve as thekey for all entities in the hierarchy.


61/69

61

Example: Subclasses

name is Key for Beers

Beers

Ales

isa

name manf

color

Notations of Inheritance


62/69

Notations of Inheritance

62


63/69

63

E/R Vs. Object-Oriented Subclasses

In OO, objects are in one class only.

Subclasses inherit from superclasses.

In contrast, E/R entities haverepresentatives in all subclasses towhich they belong.

Rule: if entity eis represented in a subclass,then eis represented in the superclass (andrecursively up the tree).


64/69

64

Example: Representatives of

Entities

Beers

Ales

isa

name manf

color

Petes Ale


65/69

65

Subclasses: Three Approaches

1. Object-oriented: One relation per subset ofsubclasses, with all relevant attributes.

2. Use nulls: One relation; entities have NULLin attributes that dont belong to them.

3. E/R style: One relation for each subclass:

Key attribute(s). Attributes of that subclass.


66/69

66

Example: Subclass -> Relations

Beers

Ales

isa

name manf

color


67/69

67

Object-Oriented

name manf

Bud Anheuser-Busch

Beers

name manf colorSummerbrew Petes dark

Ales

Good for queries like find thecolor of ales made by Petes.


68/69

68

E/R Style

name manfBud Anheuser-BuschSummerbrew Petes

Beers

name colorSummerbrew dark

Ales

Good for queries likefind all beers (includingales) made by Petes.


69/69

Using Nulls

name manf color

Bud Anheuser-Busch NULLSummerbrew Petes darkBeers

Saves space unless there are lotsof attributes that are usually NULL.

notes of awesome

Documents