Conceptual and Logical Design

Size: px

Start display at page:

Download "Conceptual and Logical Design"

Joseph Mills
6 years ago
Views:

1 Introduction to Database Systes Mapping ER Models to Relational Scheas Werner Nutt Conceptual and Logical Design Conceptual Model: PRODUCT BUYS PERSON price ssn Relational Model: 2

2 Mapping an E-R Diagra to a Relational Schea We cannot store date in an ER schea (there are no ER database anageent systes) We have to translate our ER schea into a relational schea What does translation ean? 3 Translation: Principles Maps ER scheas to relational scheas ER instances to relational instances Ideally, the apping should be one-to-one in both directions not lose any inforation Difficulties: what to do with ER-instances that have identical attribute values, but consist of different entities? in which way do we want to preserve inforation? 4

3 Mapping Entity Types to Relations For every entity type create a relation Every atoic attribute of the entity type becoes a relation attribute Coposite attributes: include all the atoic attributes Derived attributes are not included (but reeber their derivation rules) Relation instances are subsets of the cross product of the doains of the attributes Attributes of the entity key ake up the priary key of the relation faily no. of students courseno equip COURSE subject 5 Mapping Entity Types to Relations (cntd.) faily no. of students courseno equip COURSE subject (,, faily) COURSE (courseno, subject, equip) 6

4 Mapping Many:any Relationship Types to Relations Create a relation with the following set of attributes: priary keys of each entity type participating in the relationship N (degree of relationship) U priary_key(e i ) U {a,,a M } i= attributes of the relationship type (if any) faily labark no. of students courseno equip ENROLLED COURSE exaark subject 7 Mapping Many:any Relationship Types to Relations (cntd.) faily labark no. of students courseno equip ENROLLED COURSE exaark subject ENROL(, courseno, labark, exaark) Foreign Key ENROL() references () Foreign Key ENROL(courseno) references COURSE(courseno) 8

5 Mapping Many:one Relationship Types to Relations faily slot roono TUTOR Idea: Post the priary key Given E at the any end of relationship and E2 at the one end of the relationship, add inforation to the relation for E The priary key of the entity at the one end (the deterined entity) becoes a foreign key in the entity at the any end (the deterining entity). Include any relationship attributes with the foreign key entity E U priary_key(e2) U {a,,a n } relation for entity E priary key for E2, is now a foreign key to E2 attributes on the relationship type (if any) 9 Mapping Many:one Relationship Types to Relations: Exaple faily slot roono TUTOR The relation (,, faily) is extended to (,, faily, tutor, roono, slot) and the constraint Foreign Key (tutor,roono) references (,roono) 0

6 Mapping one:any Relationship Types to Relations (cntd.) faily tutor roono slot s fred jones bush B s2 ary brown kahn IT206 2B s3 sue sith goble A s4 fred bloggs goble 2.82 A s5 peter jones zobel B s6 jill peters kahn IT206 2A roono kahn IT206 bush 2.26 goble 2.82 zobel 2.34 watson IT22 woods IT204 capon A4 lindsey 2.0 barringer 2.25 Mapping One:any Relationship Types to Relations (cntd.) faily slot roono TUTOR Another Idea: If the relationship type is optional to both entity types, and an instance of the relationship is rare, and there are any attributes on the relationship then create a new relation with the set of attributes: priary_key(e) U priary_key(e2) U {a,,a } priary key for E, is now a foreign key to E; also the PK for this relation priary key for E2, is now a foreign key to E2 attributes on the relationship type (if any) 2

7 Mapping One:any Relationship Types to Relations (cntd.) faily slot roono TUTOR TUTOR(, staff, rono, slot) and Copare with the apping of any:any relationship types! Foreign key TUTOR() references () Foreign key TUTOR(staff, roono) references (, roono) 3 Mapping One:any Relationship Types to Relations (cntd.) faily s fred jones s2 ary brown s3 sue sith s4 fred bloggs s5 peter jones s6 jill peters TUTOR tutor roono slot s bush B s2 kahn IT206 2B s3 goble A s4 goble 2.82 A s5 zobel B s6 kahn IT206 2A roono kahn IT206 bush 2.26 goble 2.82 zobel 2.34 watson IT22 woods IT204 capon A4 lindsey 2.0 barringer

8 Optional Participation of the Deterined Entity ( one end ) A student entity instance ust participate in a relationship instance of REG A school entity instance need not participate in a relationship instance of REG faily hons faculty REG SCHOOL SCHOOL (hons, faculty) (,, faily,??? ) 5 Optional Participation of the Deterined Entity faily hons s fred jones ca s2 ary brown cis s3 sue sith cs s4 fred bloggs ca s5 peter jones cs s6 jill peters ca SCHOOL hons faculty ac accountancy is inforation systes cs coputer science ce coputer science i edicine a atheatics hons can t be NULL because it is andatory for a student to be registered for a school not null constraint No student is registered for i, so i doesn t occur as a foreign key value (but that s no proble) 6

9 Optional Participation of the Deterinant Entity ( any end ) faily slot roono TUTOR A student entity instance need not participate in a relationship instance of TUTOR A staff entity instance ust participate in a relationship instance of TUTOR 7 Optional Participation of the Deterinant Entity. (,, faily, tutor, roono, slot) (, roono) Integrity constraint: π,roono \ π tutor,roono = 2. (,, faily) (, roono) TUTOR(, tutor, roono, slot) Do we also need an integrity constraint? 8

10 Optional Participation of the Deterinant Entity (cntd.) faily tutor roono slot s fred jones bush B s2 ary brown kahn IT206 2B s3 sue sith goble A s4 fred bloggs goble 2.82 A s5 peter jones zobel B s6 jill peters kahn IT206 2A roono kahn IT206 bush 2.26 goble 2.82 zobel 2.34 watson IT22 woods IT204 capon A4 lindsey 2.0 barringer Mapping One:one Relationship Types to Relations year roono YEAR YEARTUTOR Post the priary key of one of the entity types into the other entity type as a foreign key, including any relationship attributes with it or Merge the entity types together Which constraint holds in this case? YEAR year yeartutor zobel 2 bush 3 capon roono year kahn IT206 NULL bush goble 2.82 NULL zobel 2.34 watson IT22 NULL woods IT204 NULL capon A4 3 lindsey 2.0 NULL barringer 2.25 NULL 20

11 Multi-Valued Attributes For each ulti-valued attribute of E i, create a relation with the attributes priary_key(e i ) U ulti-valued attribute The priary key coprises all attributes faily faily s fred jones s2 ary brown contact _CONTACT contact s Mr. Jones s Mrs Jones s2 Bill Brown s2 Mrs Jones s2 Billy-Jo Woods 2 Mapping Roles and Recursive Relationships roono appraiser appraisee APPRAISAL How can the entity appear in both of its roles? (, roono, appraiser, approono ) 22

12 Multiple Relationships between Entity Types. Treat each relationship type separately 2. Represent distinct relationships by different foreign keys drawing on the sae relation SUPERVISE faily roono EXAMINE (, roono) (,, faily,??? ) (,, faily,??? ) EXAMINER(??? ) SUPERVISOR(??? ) EXAMINER-SUPERVISOR(??? ) 23 Non-binary Relationship roono faily p TUTORS slot courseno equip n COURSE subject COURSE(courseno, subject, equip) (,, faily) (staff, roono) TUTORS(??? ) 24

13 Weak Entities Strong entity type Identifying entity for ORDER Identifying entity for LINE_ITEM CUSTOMER CUST-ORDER c- address Weak entity type Identifying entity for LINE_ITEM ORDER orderid date ORDER-MAKEUP Weak entity type LINE_ ITEM lineno quantity 25 Mapping Weak Entities to Relations Create a relation with the attributes: n priary_key(e 0 ) U U discriinator(e i ) U {a,,a n } i= Priary key of identifying strong entity type Discriinators of identifying weak entity types Attributes of the weak entity type CUSTOMER c- address CUST-ORDER ORDER orderid date 26

14 Association Entity Types An entity type that represents a relationship type: faily courseno equip COURSE subject STUD_ENROL ENROL COURSE_ENROL labark exaark 27 Association Entity Types We have: COURSE(courseno, subject, equip) (,, faily) faily courseno equip COURSE subject STUD_ENROL ENROL COURSE_ENROL labark exaark Then: ENROL(courseno,, labark, exaark) 28

15 Translation of the University Diagra faily REG hons SCHOOL faculty (,, faily, hons, tutor, tutorroo, slot, year) labark YEARREG year YEAR ENROL(, courseno, labark,exaark) ENROL TUTOR COURSE(courseno, subject, equip) exaark courseno subject n COURSE equip slot TEACH YEARTUTOR n roono appraisee appraiser APPRAISAL (lecturer,roono, appraiser, approo) TEACH(courseno, lecturer,lecroo ) YEAR(year, yeartutor, yeartutorroo) SCHOOL(hons, faculty) 29 Exercise: Supervision of PhD Students A database needs to be developed that keeps track of PhD students: For each student store the and atriculation nuber. Matriculation nubers are unique. Each student has exactly one address. An address consists of street, town and post code, and is uniquely identified by this inforation. For each lecturer store the, staff ID and office nuber. Staff ID's are unique. Each student has exactly one supervisor. A staff eber ay supervise a nuber of students. The date when supervision began also needs to be stored. 30

16 Exercise: Supervision of PhD Students For each research topic store the title and a short description. Titles are unique. Each student can be supervised in only one research topic, though topics that are currently not assigned also need to be stored in the database. Tasks: a) Design an entity relationship diagra that covers the requireents above. Do not forget to include cardinality and participation constraints. b) Based on the ER-diagra fro above, develop a relational database schea. List tables with their attributes. Identify keys and foreign keys. 3 Translating of Hierarchies: Options To store inforation about these classes, We have to define appropriate relations. faily For each relation, we have to define: set of attributes priary key In principle, there are three options: year d thesis title UNDERGRADUATE POSTGRADUATE A. Create a relation for each entity type in the schea, i.e., for both, superclass and subclasses B. Create only relations for subclasses C. Create only one relation, for the superclass 32

17 Translation into Relations: Option A. Create a relation for the superclass 2. For each subclass, create a relation over the set of attributes priary_key(superclass) U attributes of subclass The key for each subclass relation is: priary_key(superclass) Inclusion constraint (foreign keys): π key (superclass) π key (subclass i ) Covering constraint ( n = nuber of subclasses): n U π key (subclass i ) π key (superclass) i= Disjointness constraint: π key (subclas s i ) π key (subclass j ) = for i j year UNDERGRADUATE faily d thesis title POSTGRADUATE 33 Translation into Relations: Option B Create only relations for subclasses. Each subclass becoes a relation over the set of attributes: attributes of superclass U attributes of subclass The key for each subclass relation is: priary_key(superclass) Works only if coverage is total and disjoint Partial coverage: entities that are not in a subclass are lost faily Overlapping classes: redundancy Recovery of the superclass: OUTER UNION on the subclass relations year d thesis title In SQL: UNION JOIN, Outer join on the coon attributes UNDERGRADUATE POSTGRADUATE 34

18 Translation into Relations: Option C Create a single relation over the set of attributes n attributes of superclass U attributes of subclass i U { class } i= The key is: priary_key(superclass) faily Drawback: any not-applicable nulls Benefit: No need for joins Disjoint coverage: one attribute class which indicates the subclass the tuple represents Overlapping coverage: class has to represent a set of classes Partial coverage: class is null entity is fro superclass year UNDERGRADUATE d thesis title POSTGRADUATE 35 Applying the Three Translations (Overlapping Coverage) payroll no A. (payrollno,, lengthofservice) ACADEMIC(payrollno, level) TECHNICAL(payrollno, project) ADMIN(payrollno, grade) length of service level o ACADEMIC TECHNICAL project grade ADMIN B. ACADEMIC(payrollno,, lengthofservice, level) TECHNICAL(payrollno,, lengthofservice, project) ADMIN(payrollno,, lengthofservice, grade) C. (payrollno,, lengthofservice, level, project, grade, class, class2, class3) or (payrollno,, lengthofservice, level, project, grade, class) 36 class = powerset of classes

19 Specialisation Lattice with Shared Subclass payroll no Staff dept d no_hours title d task Manager salary Hourly Staff Acadeic Technical Adin Salaried Staff expertise phone Adin Manager Exercise: For each of the approaches A, B, C, decide Which tables need to be created? Which are the attributes? And which are their possible values? 37 References In preparing these slides I have used several sources. The ain ones are the following: Books: A First Course in Database Systes, by J. Ullan and J. Wido Fundaentals of Database Systes, by R. Elasri and S. Navathe Slides fro Database courses held by the following people: Enrico Franconi (Free University of Bozen-Bolzano) Carol Goble and Ian Horrocks (University of Manchester) 38

CS530 Database Architecture Models. Database Model. Prof. Ian HORROCKS. Dr. Robert STEVENS. and Design The Relational

02 - The Relational Database Model CS530 Database Architecture Models and Design Prof. Ian HORROCKS Dr. Robert STEVENS In this Section Topics Covered The basics of the relational model in the context of