Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide
  • 1
  • 2
  • 3
  • 4
  • 9
  • 16
  • 5
  • 6
  • 6
  • 7
  • 7
  • 13
  • 14
  • 3 The slides for this text are organized into several modules. Each lecture contains about enough material for a 1.25 hour class period. (The time estimate is very approximate--it will vary with the instructor, and lectures also differ in length; so use this as a rough guideline.) This covers Lectures 1 and 2 (of 6) in Module (5). Module (1): Introduction (DBMS, Relational Model) Module (2): Storage and File Organizations (Disks, Buffering, Indexes) Module (3): Database Concepts (Relational Queries, DDL/ICs, Views and Security) Module (4): Relational Implementation (Query Evaluation, Optimization) Module (5): Database Design (ER Model, Normalization, Physical Design, Tuning) Module (6): Transaction Processing (Concurrency Control, Recovery) Module (7): Advanced Topics
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 7
  • 8
  • 15
  • (.ppt)

    1. 1. The Relational Model Lecture 3 INFS614, Fall 2008
    2. 2. Relational Model <ul><li>Relational Model = Structure + Operations </li></ul><ul><ul><li>Structure: Relations (or Tables) </li></ul></ul><ul><ul><li>Operations: Relational Algebra, SQL. </li></ul></ul><ul><li>Most widely implemented model. </li></ul><ul><ul><li>Vendors: IBM DB2, Microsoft SQL Server, Oracle, etc. </li></ul></ul><ul><li>Our design+implementation approach: </li></ul><ul><ul><li>Step 1: ER design (ERD) </li></ul></ul><ul><ul><li>Step 2: Translate to Relational (Relational Schema) </li></ul></ul><ul><ul><li>Step 3: Querying over the relational model </li></ul></ul>
    3. 3. Relational Database: Definitions <ul><li>Relational database : a set of relations </li></ul><ul><li>Relation: made up of 2 parts: </li></ul><ul><ul><li>Instance : a table , with rows and columns. #Rows = cardinality , #fields = degree / arity. </li></ul></ul><ul><ul><li>Schema : specifies name of relation, plus name and type of each column. </li></ul></ul><ul><ul><ul><li>E.G. Students( sid : string, name : string, login : string, age : integer, gpa : real). </li></ul></ul></ul><ul><li>We can think of a relation as a set of rows or tuples (i.e., all rows are distinct). </li></ul>
    4. 4. Example: Instance of Students Relation <ul><li>Cardinality = 3, degree = 5, all rows distinct; </li></ul><ul><li>Do all columns in a relation instance have to </li></ul><ul><li>be distinct? </li></ul><ul><li>The order in which the rows are listed is not important; </li></ul>
    5. 5. Another Example: Employees Relation <ul><li>Employees Schema: </li></ul><ul><li>Employees ( ssn :integer, name :string, rank :char, salary :float) </li></ul><ul><li>An instance of Employees : </li></ul>
    6. 6. Example: Employees Relation (Contd.) <ul><li>An instance of Employees = { </li></ul><ul><li><633909767, Richard Boon, A, 75689.09>, </li></ul><ul><ul><ul><li><674627883,Adolfo Laurenti, B, 67890.00>, <193838904,Will Smith,C,50000.00>,…} </li></ul></ul></ul>Set of tuples (or rows)
    7. 7. Relational Database : Definitions <ul><li>Instance : a set of tuples of the relation </li></ul><ul><ul><li>A tuple : < a 1 :d 1 , …,a n :d n >, </li></ul></ul><ul><ul><ul><li>a j is an attribute name, </li></ul></ul></ul><ul><ul><ul><li>d j is the value of the attribute a j , </li></ul></ul></ul><ul><ul><ul><li>d j either belongs to Domain(a j ) or is NULL </li></ul></ul></ul><ul><ul><li>An instance of Employees = { </li></ul></ul><ul><ul><li><ssn:633909767,name:Richard Boon, rank:A, salary:75689.09>, </li></ul></ul><ul><ul><li><ssn:674627883,name:Adolfo Laurenti, rank:B, salary:67890.00>, </li></ul></ul><ul><ul><li><ssn:193838904,name:Will Smith, rank:C, salary:50000.00>, </li></ul></ul><ul><ul><ul><li>… </li></ul></ul></ul><ul><ul><li>} </li></ul></ul>
    8. 8. Relational Database: Definitions <ul><li>Relational database : a set of relations; </li></ul><ul><li>Relational database schema : the collection of schemas for the relations in the database; </li></ul>
    9. 9. Example: A Company Database Schema A First Schema : Employees ( ssn :integer, name :string, rank :integer, salary :float) Projects ( pid :integer, pname :string, budget :float) Location ( address :string, capacity :integer) Departments ( did :integer, dname :string, budget :float) Manages( ssn : integer, did : integer, since :date) Reports_To( ssnSubordinate : integer, ssnSupervisor :integer) Works_for( ssn : integer, pid : integer, hours :float) Works_in( ssn : integer, did : integer, address :string)
    10. 10. Relational Query Languages <ul><li>A major strength of the relational model: supports simple, powerful querying of data. </li></ul><ul><li>Queries can be written intuitively, and the DBMS is responsible for efficient evaluation. </li></ul><ul><ul><li>The key: precise semantics for relational queries. </li></ul></ul><ul><ul><li>Allows the optimizer to extensively re-order operations, and still ensure that the answer does not change. </li></ul></ul>
    11. 11. The SQL Query Language <ul><li>Developed by IBM (system R) in the 1970s </li></ul><ul><li>Need for a standard since it is used by many vendors </li></ul><ul><li>Standards: </li></ul><ul><ul><li>SQL-86 </li></ul></ul><ul><ul><li>SQL-89 (minor revision) </li></ul></ul><ul><ul><li>SQL-92 (major revision, current standard) </li></ul></ul><ul><ul><li>SQL-99 (major extensions) </li></ul></ul>
    12. 12. Creating Relations in SQL <ul><li>Creates the Students relation. Observe that the type (domain) of each field is specified, and enforced by the DBMS whenever tuples are added or modified. </li></ul><ul><li>As another example, the Enrolled table holds information about courses that students take. </li></ul>CREATE TABLE Students (sid CHAR(20) , name CHAR(20) , login CHAR(10), age INTEGER , gpa REAL ) CREATE TABLE Enrolled (sid CHAR(20) , cid CHAR(20) , grade CHAR (2))
    13. 13. Adding and Deleting Tuples <ul><li>We can insert a single tuple using: </li></ul>INSERT INTO Students (sid, name, login, age, gpa) VALUES (53688, ‘Smith’, ‘smith@ee’, 18, 3.2) <ul><li>Can delete all tuples satisfying some condition (e.g., name = Smith): </li></ul>DELETE FROM Students S WHERE S.name = ‘Smith’ <ul><li>Powerful variants of these commands are available; more later! </li></ul>
    14. 14. Querying Relational Data <ul><li>To find all 18 year old students, we can write: </li></ul>SELECT * FROM Students S WHERE S.age=18 Instance of Students:
    15. 15. Querying Relational Data (Contd.) <ul><li>The result is: </li></ul>SELECT * FROM Students S WHERE S.age=18 <ul><li>To find just names and logins, replace the first line: </li></ul>SELECT S.name, S.login
    16. 16. Updating Tuples <ul><li>Can update tuples using: </li></ul>UPDATE Students S SET S.age = S.age + 1, S.gpa = S.gpa -1 WHERE S.sid = 53688
    17. 17. Querying Multiple Relations <ul><li>What does the following query compute? </li></ul>SELECT S.name, E.cid FROM Students S, Enrolled E WHERE S.sid=E.sid AND E.grade=“A”
    18. 18. Querying Multiple Relations we get: Instance of Students: Instance of Enrolled:
    19. 19. Destroying and Altering Relations <ul><li>Destroys the relation Students. The schema information and the tuples are deleted. </li></ul>DROP TABLE Students <ul><li>The schema of Students is altered by adding a new field; every tuple in the current instance is extended with a null value in the new field. </li></ul>ALTER TABLE Students ADD COLUMN firstYear: INTEGER
    20. 20. Integrity Constraints (ICs) <ul><li>IC: condition that must be true for any instance of the database; e.g., domain constraints. </li></ul><ul><ul><li>ICs are specified when schema is defined. </li></ul></ul><ul><ul><li>ICs are checked when relations are modified. </li></ul></ul><ul><li>A legal instance of a relation is one that satisfies all specified ICs. </li></ul><ul><ul><li>DBMS should not allow illegal instances. </li></ul></ul><ul><li>If the DBMS checks ICs, stored data is more faithful to real-world meaning. </li></ul><ul><ul><li>Avoids data entry errors, too! </li></ul></ul>
    21. 21. Primary Key Constraints <ul><li>A set of fields is a (candidate) key for a relation if : </li></ul><ul><ul><li>1. No two distinct tuples can have same values in all key fields, and </li></ul></ul><ul><ul><li>2. This is not true for any subset of the key. </li></ul></ul><ul><ul><li>Part 2 false? A superkey . </li></ul></ul><ul><ul><li>If there’s >1 candidate keys for a relation, one of the keys is chosen (by DBA) to be the primary key . </li></ul></ul><ul><li>E.g., sid is a key for Students. (What about name ?) The set { sid, gpa } is a superkey. </li></ul>
    22. 22. Primary and Candidate Keys in SQL <ul><li>Possibly many candidate keys (specified using UNIQUE ), one of which is chosen as the primary key . </li></ul>CREATE TABLE Enrolled (sid CHAR (20) cid CHAR(20) , grade CHAR (2), PRIMARY KEY (sid,cid) ) <ul><li>“ For a given student and course, there is a single grade.” vs. “Students can take only one course, and receive a single grade for that course; further, no two students in a course receive the same grade.” </li></ul><ul><li>Used carelessly, an IC can prevent the storage of database instances that arise in practice! </li></ul>CREATE TABLE Enrolled (sid CHAR (20) cid CHAR(20) , grade CHAR (2), PRIMARY KEY (sid), UNIQUE (cid, grade) )
    23. 23. Foreign Keys, Referential Integrity <ul><li>In addition to Students we have a second relation: </li></ul><ul><li>Enrolled( sid : string, cid : string, grade : string) </li></ul><ul><li>Only students listed in the Students relation should be allowed to enroll for courses . </li></ul>Enrolled Students
    24. 24. Foreign Keys, Referential Integrity <ul><li>Foreign key : Set of fields in one relation that is used to `refer’ to a tuple in another relation. (Must correspond to primary key of the second relation.) Like a `logical pointer’. </li></ul><ul><li>E.g. sid is a foreign key referring to Students : </li></ul><ul><ul><li>Enrolled( sid : string, cid : string, grade : string) </li></ul></ul><ul><ul><li>If all foreign key constraints are enforced, referential integrity is achieved, i.e., no dangling references. </li></ul></ul><ul><ul><li>Can you name a data model w/o referential integrity? </li></ul></ul><ul><ul><ul><li>Links in HTML! </li></ul></ul></ul>
    25. 25. Foreign Keys, Referential Integrity <ul><li>Another Example : </li></ul><ul><ul><li>Only employees in the Employees Relation should be allowed to be managers: </li></ul></ul><ul><ul><ul><li>ssn is a Foreign Key respect to Employees </li></ul></ul></ul><ul><ul><li>Only projects in the Project Relation should be allowed to be managed : </li></ul></ul><ul><ul><ul><li>pid is a Foreign Key respect to Projects </li></ul></ul></ul>Employees Managers Projects
    26. 26. Foreign Keys in SQL <ul><li>Only students listed in the Students relation should be allowed to enroll for courses. </li></ul>CREATE TABLE Enrolled (sid CHAR (20), cid CHAR(20) , grade CHAR (2), PRIMARY KEY (sid,cid), FOREIGN KEY (sid) REFERENCES Students ) Enrolled Students
    27. 27. Foreign Keys, Referential Integrity <ul><li>A Foreign Key must correspond to the primary key of the referenced relation </li></ul><ul><li>A Foreign Key states a Referential IC between two relations : a tuple in one relation that refers to another must refer to an existing tuple in that relation. </li></ul><ul><li>Referential Integrity Constraints is used to maintain the consistency among tuples of two related relations </li></ul>
    28. 28. Enforcing Referential Integrity <ul><li>Consider Students and Enrolled; sid in Enrolled is a foreign key that references Students. </li></ul><ul><li>What should be done if an Enrolled tuple with a non-existent student id is inserted? ( Reject it! ) </li></ul><ul><li>What should be done if a Students tuple is deleted? </li></ul><ul><ul><li>Also delete all Enrolled tuples that refer to it. </li></ul></ul><ul><ul><li>Disallow deletion of a Students tuple that is referred to. </li></ul></ul><ul><ul><li>Set sid in Enrolled tuples that refer to it to a default sid . </li></ul></ul><ul><ul><li>(In SQL, also: Set sid in Enrolled tuples that refer to it to a special value null , denoting `unknown’ or `inapplicable’ .) </li></ul></ul><ul><li>Similarly if primary key value of a Students tuple is updated. </li></ul>
    29. 29. Referential Integrity in SQL/92 <ul><li>SQL/92 supports all 4 options on deletes and updates. </li></ul><ul><ul><li>Default is NO ACTION ( delete/update is rejected ) </li></ul></ul><ul><ul><li>CASCADE (also delete all tuples that refer to deleted tuple) </li></ul></ul><ul><ul><li>SET NULL / SET DEFAULT (sets foreign key value of referencing tuple) </li></ul></ul>CREATE TABLE Enrolled (sid CHAR (20), cid CHAR(20) , grade CHAR (2), PRIMARY KEY (sid,cid), FOREIGN KEY (sid) REFERENCES Students ON DELETE CASCADE ON UPDATE NO ACTION )
    30. 30. Where do ICs Come From? <ul><li>ICs are based upon the semantics of the real-world enterprise that is being described in the database relations. </li></ul><ul><li>We can check a database instance to see if an IC is violated, but we can NEVER infer that an IC is true by looking at an instance. </li></ul><ul><ul><li>An IC is a statement about all possible instances! </li></ul></ul><ul><ul><li>From example, we know name is not a key, but the assertion that sid is a key is given to us. </li></ul></ul><ul><li>Key and foreign key ICs are the most common; more general ICs supported too. </li></ul>
    31. 31. Logical DB Design: ER to Relational <ul><li>Entity sets to tables. </li></ul>CREATE TABLE Employees (ssn CHAR (11), name CHAR (20), lot INTEGER , PRIMARY KEY (ssn) ) Employees ssn name lot
    32. 32. Relationship Sets to Tables <ul><li>In translating a relationship set to a relation, attributes of the relation must include: </li></ul><ul><ul><li>Keys for each participating entity set (as foreign keys). </li></ul></ul><ul><ul><ul><li>This set of attributes forms a superkey for the relation. </li></ul></ul></ul><ul><ul><li>All descriptive attributes. </li></ul></ul>CREATE TABLE Works_In( ssn CHAR (11), did INTEGER , since DATE , PRIMARY KEY (ssn, did), FOREIGN KEY (ssn) REFERENCES Employees, FOREIGN KEY (did) REFERENCES Departments )
    33. 33. Translating Ternary Relationship Set Works_In0( ssn : integer, did :integer, address :string) CREATE TABLE Works_In0 (ssn CHAR(11), did INTEGER , address CHAR(60), PRIMARY KEY (ssn, did, address), FOREIGN KEY (ssn) REFERENCES Employees, FOREIGN KEY (did) REFERENCES Departments, FOREIGN KEY (address) REFERENCES Locations) Works-In0 Departments dname did Locations address Capacity dbudget Employees salary name ssn rank
    34. 34. Review: Key Constraints <ul><li>Each dept has at most one manager, according to the key constraint on Manages. </li></ul>Translation to relational model? Many-to-Many 1-to-1 1-to Many Many-to-1 budget did Departments dname since lot name ssn Manages Employees
    35. 35. Translating ER Diagrams with Key Constraints <ul><li>Map relationship to a table: </li></ul><ul><ul><li>Note that did is the key now! </li></ul></ul><ul><ul><li>Separate tables for Employees and Departments. </li></ul></ul><ul><li>Since each department has a unique manager, we could instead combine Manages and Departments. </li></ul>CREATE TABLE Manages( ssn CHAR(11) , did INTEGER , since DATE , PRIMARY KEY (did) , FOREIGN KEY (ssn) REFERENCES Employees, FOREIGN KEY (did) REFERENCES Departments ) CREATE TABLE Dept_Mgr( did INTEGER, dname CHAR(20), budget REAL, ssn CHAR(11) , since DATE , PRIMARY KEY (did), FOREIGN KEY (ssn) REFERENCES Employees )
    36. 36. Review: Participation Constraints <ul><li>Does every department have a manager? </li></ul><ul><ul><li>If so, this is a participation constraint : the participation of Departments in Manages is said to be total (vs. partial ) . </li></ul></ul><ul><ul><ul><li>Every did value in Departments table must appear in a row of the Manages table (with a non-null ssn value!) </li></ul></ul></ul>lot name dname budget did since name dname budget did since Manages since Departments Employees ssn Works_In
    37. 37. Participation Constraints in SQL <ul><li>We can capture participation constraints involving one entity set in a binary relationship, but little else (without resorting to CHECK constraints). </li></ul>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 )
    38. 38. Participation Constraints <ul><li>Is it possible to express this participation constraint using only key and foreign key constraints? </li></ul><ul><li>Works_for( ssn : integer, pid : integer, hours : float) </li></ul>CREATE TABLE Works_for (ssn INTEGER, pid INTEGER, hours float, PRIMARY KEY (ssn,pid), FOREIGN KEY (ssn) REFERENCES Employees, FOREIGN KEY (pid) REFERENCES Projects) NO Works_For Projects pid pname hours Employees salary name ssn rank
    39. 39. Review: Weak Entities <ul><li>A weak entity can be identified uniquely only by considering the primary key of another ( owner ) entity. </li></ul><ul><ul><li>Owner entity set and weak entity set must participate in a one-to-many relationship set (1 owner, many weak entities). </li></ul></ul><ul><ul><li>Weak entity set must have total participation in this identifying relationship set. </li></ul></ul>lot name age pname Dependents Employees ssn Policy cost
    40. 40. Translating Weak Entity Sets <ul><li>Weak entity set and identifying relationship set are translated into a single table. </li></ul><ul><ul><li>When the owner entity is deleted, all owned weak entities must also be deleted. </li></ul></ul>CREATE TABLE Dep_Policy ( pname CHAR(20) , age INTEGER , cost REAL , ssn CHAR(11) , PRIMARY KEY (pname, ssn), FOREIGN KEY (ssn) REFERENCES Employees, ON DELETE CASCADE )
    41. 41. Review: ISA Hierarchies <ul><li>Overlap constraints : Can Joe be an Hourly_Emps as well as a Contract_Emps entity? ( Allowed/disallowed ) </li></ul><ul><li>Covering constraints : Does every Employees entity also have to be an Hourly_Emps or a Contract_Emps entity? (Yes/no) </li></ul>Contract_Emps name ssn Employees lot hourly_wages ISA Hourly_Emps contractid hours_worked <ul><li>As in C++, or other PLs, attributes are inherited. </li></ul><ul><li>If we declare A ISA B, every A entity is also considered to be a B entity. </li></ul>
    42. 42. Translating ISA Hierarchies to Relations <ul><li>General approach: </li></ul><ul><ul><li>3 relations: Employees, Hourly_Emps and Contract_Emps. </li></ul></ul><ul><ul><ul><li>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). </li></ul></ul></ul><ul><ul><ul><li>Queries involving all employees easy, those involving just Hourly_Emps require a join to get some attributes. </li></ul></ul></ul><ul><li>Alternative: Just Hourly_Emps and Contract_Emps. </li></ul><ul><ul><li>Hourly_Emps : ssn , name, lot, hourly_wages, hours_worked. </li></ul></ul><ul><ul><li>Each employee must be in one of these two subclasses . </li></ul></ul>
    43. 43. Translating ISA Hierarchies to Relations <ul><li>General approach: </li></ul>CREATE TABLE Hourly_Emps ( hourly_wages REAL, hours_worked REAL, ssn CHAR(11), PRIMARY KEY (ssn), FOREIGN KEY (ssn) REFERENCES Employees, ON DELETE CASCADE ) <ul><li>Similarly for Contract_Emps TABLE </li></ul>
    44. 44. Translating Aggregations Monitors until Sponsors Departments dbudget Projects dname did pid pname since Employees salary name ssn rank
    45. 45. Translating Aggregations Sponsors ( did, pid , since) Monitors ( ssn , did , pid ,until) CREATE TABLE Sponsors (did INTEGER, pid INTEGER , since DATE, PRIMARY KEY (did,pid), FOREIGN KEY (did) REFERENCES Departments, FOREIGN KEY (pid) REFERENCES Projects) CREATE TABLE Monitors (ssn INTEGER, did INTEGER, pid INTEGER , until DATE, PRIMARY KEY (ssn, did, pid), FOREIGN KEY (did,pid) REFERENCES Sponsors, FOREIGN KEY (ssn) REFERENCES Employees)
    46. 46. Translating Aggregations <ul><li>If: Every sponsored project has a monitor, and </li></ul><ul><li>the attribute “since” is not required for Sponsors …. </li></ul>Therefore, we can omit the Sponsors relation Every possible instance of the Sponsors relationship is obtained by looking at the set of pairs < pid , did > in the relation Monitors
    47. 47. Review: Binary vs. Ternary Relationships <ul><li>If each policy is owned by just 1 employee: </li></ul><ul><ul><li>Key constraint on Policies would mean policy can only cover 1 dependent! </li></ul></ul><ul><li>What are the additional constraints in the 2nd diagram? </li></ul>age pname Dependents Covers age pname Dependents Purchaser Bad design Better design name Employees ssn lot Policies policyid cost Beneficiary policyid cost Policies name Employees ssn lot
    48. 48. Binary vs. Ternary Relationships (Contd.) <ul><li>The key constraints allow us to combine Purchaser with Policies and Beneficiary with Dependents. </li></ul><ul><li>Participation constraints lead to NOT NULL constraints. </li></ul>CREATE TABLE Policies ( policyid INTEGER , cost REAL , ssn CHAR(11) NOT NULL , PRIMARY KEY (policyid), FOREIGN KEY (ssn) REFERENCES Employees, ON DELETE CASCADE ) CREATE TABLE Dependents ( pname CHAR(20) , age INTEGER , policyid INTEGER , PRIMARY KEY (pname, policyid), FOREIGN KEY (policyid) REFERENCES Policies, ON DELETE CASCADE )
    49. 49. Relational Model: Summary <ul><li>A tabular representation of data. </li></ul><ul><li>Simple and intuitive, currently the most widely used. </li></ul><ul><li>Integrity constraints can be specified by the DBA, based on application semantics. DBMS checks for violations. </li></ul><ul><ul><li>Two important ICs: primary and foreign keys </li></ul></ul><ul><ul><li>In addition, we always have domain constraints. </li></ul></ul><ul><li>Powerful and natural query languages exist. </li></ul><ul><li>Rules to translate ER to relational model </li></ul>