Transcript Conceptual Design Using the ER Model

Conceptual Design Using the
Entity-Relationship (ER) Model
Module 2, Lectures 3
Database Management Systems,
1
Database Design Process

Requirements analysis
– What data, what applications, what most frequent
operations,…

Conceptual database design
– High level description of the data and the
constraint
– This step can use ER or similar high level models

Logical database design
– Convert database design into a database schema,
e.g. relational db schema
Database Management Systems,
2
Database Design Process (cont.)

Schema refinement
– Analyze the the collection of the data for potential
problems and refine it

Physical database design
– Ensure that the design meets the performance
requirements, based on used indexation, etc.

Security design
– Identify different user groups with different roles,
so that data protection is enforced accordingly.
Database Management Systems,
3
ER Model Basics


ssn
Entity: Real-world object
distinguishable from
other objects.
– An entity is described (in DB)
using a set of attributes.
Entity Set: A collection of similar
entities. E.g., all employees.
name
lot
Employees
ssn
name
lot
123-22-3666 Attishoo
48
231-31-5368 Smiley
22
131-24-3650 Smethurst 35
– All entities in an entity set have the
same set of attributes.
CREATE TABLE Employees
– Each entity set has a key, uniquely
(ssn CHAR(11),
identifies it.
name CHAR(20),
– Each attribute has a domain.
lot INTEGER,
– Can map entity set to a relation easily.
PRIMARY KEY (ssn)) 4
Database Management Systems,
ER Model Basics (Contd.)
name
ssn
lot
since
name
ssn
dname
lot
did
Employees
budget
supervisor
Employees
Works_In
Departments
subordinate
Reports_To
Relationship: Association among 2 or more entities.
E.g., Attishoo works in Pharmacy department.
 Relationship Set: Collection of similar relationships.

– An n-ary relationship set R relates n entity sets E1 ... En;
each relationship in R involves entities e1  E1, ..., en  En
 Same entity set could participate in different
relationship sets, or in different “roles” in same set.
Database Management Systems,
5
ER Model Basics (Contd.)
CREATE TABLE Works_In(
 Relationship sets can also have
ssn CHAR(1),
descriptive attributes (e.g., the
did INTEGER,
since attribute of Works_In).
since DATE,
 In translating a relationship set
PRIMARY KEY (ssn, did),
to a relation, attributes of the
FOREIGN KEY (ssn)
relation must include:
REFERENCES Employees,
FOREIGN KEY (did)
– Keys for each participating
REFERENCES Departments)
entity set (as foreign keys).
This set of attributes
forms superkey for the
relation.
– All descriptive attributes.

Database Management Systems,
ssn
123-22-3666
123-22-3666
231-31-5368
did
51
56
51
since
1/1/91
3/3/93
2/2/92
6
Key Constraints
ssn


since
name
dname
lot
did
Consider Works_In:
Manages
Employees
An employee can
work in many
departments; a dept
can have many
employees.
In contrast, each
dept has at most
one manager,
according to the
1-to Many
Many-to-1
1-to-1
key constraint on
Manages.
* Translation to relational model?
Database Management Systems,
budget
Departments
Many-to-Many
7
Translating ER Diagrams with Key Constraints



CREATE TABLE Manages(
Map relationship to a
table:
ssn CHAR(11),
did INTEGER,
– Note that did is the
key now!
since DATE,
PRIMARY KEY (did),
– Separate tables for
Employees and
FOREIGN KEY (ssn) REFERENCES Employees,
Departments.
FOREIGN KEY (did) REFERENCES Departments)
Since each department
CREATE TABLE Dept_Mgr(
has a unique manager,
did INTEGER,
we could instead
combine Manages and
dname CHAR(20),
Departments.
budget REAL,
Every dept. may not have
ssn CHAR(11),
a manager, null values
since DATE,
allowed
PRIMARY KEY (did),
FOREIGN KEY (ssn) REFERENCES Employees)
Database Management Systems,
8
Participation Constraints

Every department have a manager
– If so, this is a participation constraint: the participation of
Departments in Manages is said to be total (vs. partial).
 Every did value in Departments table must appear in a
row of the Manages table (with a non-null ssn value!)
In the following query the Null value for the ssn in
Dept_Mgr is not allowed…
 NO ACTION specification is actually the default case,
if not mentioned

– It ensures that an Employee tuple cannot be deleted while
it is pointed to by a Dept_Mgr
Database Management Systems,
9
Participation Constraints in SQL

We can capture participation constraints involving
one entity set in a binary relationship, but little else
(without resorting to CHECK constraints).
CREATE TABLE Dept_Mgr(
did INTEGER,
dname CHAR(20),
budget REAL,
ssn CHAR(11) NOT NULL,
since DATE,
PRIMARY KEY (did),
FOREIGN KEY (ssn) REFERENCES Employees,
ON DELETE NO ACTION)
Database Management Systems,
10
Participation Constraints (cont.)







Many participations cannot be captured by SQL-92
For example, take the following two relationship set
(Manages and Works_In) ER diagram.
How to ensure total participation in SQL relation
corresponding Works_In relationship?
We have to guarantee that every did value in
Departments appears in a tuple of Works_In
This tuple must also have non null values in the
foreign key fields
This cannot be achieved similar to Manages
relationship, because did cannot be taken as key for
Works_In relationship
This situation needs assertions…
Database Management Systems,
11
Participation Constraints (cont.)





Another constraint that cannot be express in SQL is
the requirement that each employee must manage at
least one department.
Such cases requires constraints that involve more than
one table.
In SQL, this is achieved using Assertions, which are
constraints that associated to multiple relations or
tables.
Assertions are imposed by the CHECK clause in SQL,
although its implementation is cumbersome.
All the entities and the relationships can be mapped
to one single relation!
Database Management Systems,
12
Participation constraints (cont.)

Two relationship set ER
since
name
ssn
dname
did
lot
Employees
Manages
budget
Departments
Works_In
since
Database Management Systems,
13
Weak Entities
A weak entity can be identified uniquely only by
considering the primary key of another (owner) entity.
 Owner entity set and weak entity set must participate
in a one-to-many relationship set.
 Weak entity set must have total participation in this
identifying relationship set.
 A weak entity always has a partial key, it can only be
uniquely defined if we take its key together with the
key of the owner entity.

Database Management Systems,
14
Weak Entities (cont.)

Employee is owner entity, Dependent is weak
entity, Policy is the relationship set, in this ER
diagram.
name
ssn
lot
Employees
Database Management Systems,
cost
Policy
pname
age
Dependents
15
Translating Weak Entity Sets

Weak entity set and identifying relationship
set are translated into a single table.
– When the owner entity is deleted, all owned weak
entities must also be deleted.
CREATE TABLE Dep_Policy (
pname CHAR(20),
age INTEGER,
cost REAL,
ssn CHAR(11) NOT NULL,
PRIMARY KEY (pname, ssn),
FOREIGN KEY (ssn) REFERENCES Employees,
ON DELETE CASCADE)
Database Management Systems,
16
name
ISA (`is a’) Hierarchies
ssn
lot
Employees
As
in C++, or other PLs,
hours_worked
hourly_wages
attributes are inherited.
ISA
contractid
If we declare A ISA B, every A
entity is also considered to be a B
Contract_Emps
Hourly_Emps
entity. (Query answers should
reflect this: unlike C++!)
 Overlap constraints: Can Joe be an Hourly_Emps as well as a
Contract_Emps entity? (Allowed/disallowed)
 Covering constraints: Does every Employees entity also have to
be an Hourly_Emps or a Contract_Emps entity? (Yes/no)
 Reasons for using ISA:
– To add descriptive attributes specific to a subclass.
– To identify entities that participate in a relationship.
Database Management Systems,
17
Translating ISA Hierarchies to Relations

General approach:
– 3 relations: Employees, Hourly_Emps and Contract_Emps.
 Hourly_Emps: Every employee is recorded in
Employees. For hourly emps, extra info recorded in
Hourly_Emps (hourly_wages, hours_worked, ssn); must
delete Hourly_Emps tuple if referenced Employees
tuple is deleted).
 Queries involving all employees easy, those involving
just Hourly_Emps require a join to get some attributes.

Alternative: Just Hourly_Emps and Contract_Emps.
– Hourly_Emps: ssn, name, lot, hourly_wages, hours_worked.
– Each employee must be in one of these two subclasses.
Database Management Systems,
18
Translating ISA Hierarchies to
Relations(cont.)
The second alternative is not applicable if
there are employees who are neither of the
subclasses.
 Also, with the second method there will be
more redundant fields, repeated in the subs.
 Also, overlap and covering constraints can
only be expressed using assertions, in SQL-92.

Database Management Systems,
19
Aggregation



Monitors has three attributes: ssn, (did,
pid), until
Used when we have to
Sponsors has three attributes (did, pid),
model a relationship
since
involving (entitity sets
Obviously, not every sponsorship appears
and) a relationship set.
Aggregation allows us to in Monitors
treat a relationship set as If the Sponsors has total participation in
an entity set for
Monitors, no separate relation is required
purposes of
for it.
participation in (other)
Monitors is a distinct relationship,
relationships.
with a descriptive attribute.
Monitors mapped to
 Also, can say that each sponsorship
table like any other
relationship set.
is monitored by at most one employee.
Database Management Systems,
20
Aggregation Example
name
ssn
lot
Employees
Monitors
until
started_on
pid
dname
pbudget
Projects
Database Management Systems,
did
Sponsors
budget
Departments
21
Conceptual Design Using the ER Model

Design choices:
– Should a concept be modelled as an entity or an attribute?
– Should a concept be modelled as an entity or a relationship?
– Identifying relationships: Binary or ternary? Aggregation?

Constraints in the ER Model:
– A lot of data semantics can (and should) be captured.
– But some constraints cannot be captured in ER diagrams.

Need for further refining the schema:
– Relational schema obtained from ER diagram is a good first
step. But ER design subjective & can’t express certain
constraints; so this relational schema may need refinement.
Database Management Systems,
22
Entity vs. Attribute
Should address be an attribute of Employees or an
entity (connected to Employees by a relationship)?
 Depends upon the use we want to make of address
information, and the semantics of the data:

If we have several addresses per employee, address must
be an entity (since attributes cannot be set-valued).
 If the structure (city, street, etc.) is important, e.g., we
want to retrieve employees in a given city, address must
be modelled as an entity (since attribute values are
atomic).

Database Management Systems,
23
Entity vs. Attribute (Contd.)


name
from
to
dname
Works_In2 does not ssn
lot
did
budget
allow an employee to
Departments
Works_In2
Employees
work in a department
for two or more periods.
Similar to the problem
of wanting to record
several addresses for an
name
dname
employee: we want to
ssn
lot
did
budget
record several values of the
Works_In3
Departments
Employees
descriptive attributes for
each instance of this
Duration
to
from
relationship.
Database Management Systems,
24
Entity vs. Relationship


First ER diagram OK if
since dbudget
name
a manager gets a
ssn
lot
did
separate discretionary
budget for each dept.
Employees
Manages2
What if a manager gets
a discretionary budget
that covers all
name
managed depts?
ssn
lot
did
– Redundancy of dbudget,
Employees
which is stored for each
dept managed by the
manager.
Misleading: suggests dbudget
apptnum
tied to managed dept.
Database Management Systems,
Manages3
dname
budget
Departments
dname
budget
Departments
since
Mgr_Appts
dbudget
25
Binary vs. Ternary Relationships
name
ssn

If each policy is
owned by just 1
employee:
– Key constraint
on Policies
would mean
policy can only
cover 1
dependent!

What are the
additional
constraints in the
2nd diagram?
Database Management Systems,
pname
lot
Employees
Dependents
Covers
Bad design
Policies
policyid
cost
name
ssn
age
pname
lot
age
Dependents
Employees
Purchaser
Better design
policyid
Beneficiary
Policies
cost
26
Binary vs. Ternary Relationships (Contd.)
CREATE TABLE Policies (
 The key
policyid INTEGER,
constraints allow cost REAL,
us to combine
ssn CHAR(11) NOT NULL,
Purchaser with
PRIMARY KEY (policyid).
Policies and
FOREIGN KEY (ssn) REFERENCES Employees,
Beneficiary with
ON DELETE CASCADE)
Dependents.
Participation
CREATE TABLE Dependents (
constraints lead to pname CHAR(20),
NOT NULL
age INTEGER,
constraints.
policyid INTEGER,
 What if Policies is
PRIMARY KEY (pname, policyid).
a weak entity set?
FOREIGN KEY (policyid) REFERENCES Policies,
ON DELETE CASCADE)
Database Management Systems,
27

Binary vs. Ternary Relationships (Contd.)
Previous example illustrated a case when 2 binary
relationships were better than a ternary relationship.
 An example in the other direction: a ternary relation
Contracts relates entity sets Parts, Departments and
Suppliers, and has descriptive attribute qty. No
combination of binary relationships is an adequate
substitute:

– S ``can-supply’’ P, D ``needs’’ P, and D ``deals-with’’ S
does not imply that D has agreed to buy P from S.
– How do we record qty?
Database Management Systems,
28
Constraints Beyond the ER Model

Functional dependencies:
– e.g., A dept can’t order two distinct parts from the same supplier.
 Can’t express this wrt ternary Contracts relationship.
– Normalization refines ER design by considering FDs.

Inclusion dependencies:
– Special case: Foreign keys (ER model can express these).
– e.g., At least 1 person must report to each manager. (Set of ssn
values in Manages must be subset of supervisor_ssn values
in Reports_To.) Foreign key? Expressible in ER model?

General constraints:
– e.g., Manager’s discretionary budget less than 10% of the
combined budget of all departments he or she manages.
Database Management Systems,
29
Summary of Conceptual Design

Conceptual design follows requirements analysis,
– Yields a high-level description of data to be stored

ER model popular for conceptual design
– Constructs are expressive, close to the way people think
about their applications.
Basic constructs: entities, relationships, and attributes
(of entities and relationships).
 Some additional constructs: weak entities, ISA
hierarchies, and aggregation.
 Note: There are many variations on ER model.

Database Management Systems,
30
Summary of ER (Contd.)

Several kinds of integrity constraints can be expressed
in the ER model: key constraints, participation
constraints, and overlap/covering constraints for ISA
hierarchies. Some foreign key constraints are also
implicit in the definition of a relationship set.
– Some of these constraints can be expressed in SQL only if
we use general CHECK constraints or assertions.
– Some constraints (notably, functional dependencies) cannot be
expressed in the ER model.
– Constraints play an important role in determining the best
database design for an enterprise.
Database Management Systems,
31
Summary of ER (Contd.)

ER design is subjective. There are often many ways to
model a given scenario! Analyzing alternatives can be
tricky, especially for a large enterprise. Common
choices include:
– Entity vs. attribute, entity vs. relationship, binary or n-ary
relationship, whether or not to use ISA hierarchies, and
whether or not to use aggregation.

Ensuring good database design: resulting relational
schema should be analyzed and refined further. FD
information and normalization techniques are
especially useful.
Database Management Systems,
32
Views


Views are computed as needed using view definition
Example
CREATE VIEW B-Students(name, sid, course)
AS SELECT S.name, S.sid, E.cid
FROM Students S, Enrolled E
WHERE S.sid=E.sid AND E.grade=‘B’



Whenever B-Students is used in a query, the view
definition is first evaluated, before using B-Students in
any other query operation.
View concept provides logical data independence, as
it can be used to mask the changes in the conceptual
schema.
Views can be defined taken security aspect into
consideration, e.g., dbadmin, user, group, etc..
Database Management Systems,
33
Updates on Views





The updateable views are the one defined on single
base tables, without aggregate operations.
Update on an updateable table implies update of the
corresponding base table as well.
A view can be dropped by DROP VIEW command.
Deleting base tables is more restrictive, as they have
to have RESTRICT or CASCADE options, concerning
integrity…
With RESTRICT, drop of a base is possible if no
reference to it exist; with CASCADE, drop of base
causes drop of all references recursively…
Database Management Systems,
34