Database Management Systems

Database Management Systems

1. Introduction Introduction; An example; Characteristics of Database approach; Actors on the screen; Workers behind the scene; Advantages of using DBMS approach; A brief history of database applications; when not to use a DBMS. Data models, schemas and instances; Three-schema architecture and data independence; Database languages and interfaces; The database system environment; Centralized and clientserver architectures; Classification of Database Management systems. 2. Entity-Relationship Model Using High-Level Conceptual Data Models for Database Design; An Example Database Application; Entity Types, Entity Sets, Attributes and Keys; Relationship types, Relationship Sets, Roles and Structural Constraints; Weak Entity Types; Refining the ER Design; ER Diagrams, Naming Conventions and Design Issues; Relationship types of degree higher than two. 3. Relational Model and Relational Algebra Relational Model Concepts; Relational Model Constraints and Relational Database Schemas; Update Operations, Transactions and dealing with constraint violations; Unary Relational Operations: SELECT and PROJECT; Relational Algebra Operations from Set Theory; Binary Relational Operations: JOIN and DIVISION; Additional Relational Operations; Examples of Queries in Relational Algebra; Relational Database Design Using ER- toRelational Mapping. 4. SQL SQL Data Definition and Data Types; Specifying basic constraints in SQL; Schema change statements in SQL; Basic queries in SQL; More complex SQL Queries. Insert, Delete and Update statements in SQL; Specifying constraints as Assertion and Trigger; Views (Virtual Tables) in SQL; Additional features of SQL; Database programming issues and techniques; Embedded SQL, Dynamic SQL; Database stored procedures and SQL/PSM. 5. Database Design Informal Design Guidelines for Relation Schemas; Functional Dependencies; Normal Forms Based on Primary Keys; General Definitions of Second and Third Normal Forms; BoyceCodd Normal Form. 6. Transaction Management The ACID Properties; Transactions and Schedules; Concurrent Execution of Transactions; Lock- Based Concurrency Control; Performance of locking; Transaction support in SQL; Introduction to crash recovery; 2PL, Serializability and Recoverability; Lock Management; Introduction to ARIES; The log; Other recovery-related structures; The write-ahead log protocol; Checkpointing; Recovering from a System Crash; Media Recovery; Other approaches and interaction with concurrency control. Text Books: 1. Elmasri and Navathe: Fundamentals of Database Systems, 5th Edition, Addison-Wesley, 2007 2. Raghu Ramakrishnan and Johannes Gehrke: Database Management

VIJAYA COLLEGE

Page 1

Database Management Systems

Systems, 3rd Edition, McGraw-Hill, 2003. Reference Book: 1. Silberschatz, Korth and Sudharshan: Data base System Concepts, 5th Edition, Mc-GrawHill, 2006.

CHAPTER 1: INTRODUCTION
DATABASE MANAGEMENT SYSTEM Data may be used to represent thing like name, telephone number, address, people name. The data can be stored using Microsoft ACCESS, EXCEL, etc. The data base is collections of data are group together to make large information. The database system is collection of programs that enables the users to create and manipulate a dbase. For example, to creates a new account in a bank and creates a new account in yahoo. The database management system is a general purpose software system, in which specifies defining, constructing and manipulating of the database. DATABASE(Dbase) MANAGEMENT SYSTEM 1. It is collection of programs that enables users to create and manipulate a dbase a. Ex: creates a new account in yahoo.com b. Ex: creates a new account in a bank 2. It is a general purpose software system 3. It specifies defining, constructing and manipulating An Example 1. In University database system, it has four files namely STUDENT, COURSE, SECTION, GRADE_REPORT schema 2. Ex: Student dbase 3. STUDENT file has four attributes Table 1. STUDENT file Student name Register no Semester Major Varaprasad 001 1 CSE Prasad 004 2 ECE 4. Ex: COURSE file has four attributes Table 2. COURSE file Course name Course no Credit Dept OS 08SCS11 3 CSE DS&A 08SCS12 4 CSE DBMS 08SCS13 4 CSE CSPA 08SCS14 4 CSE TFCS 08SCS151 3 CSE 5. Ex: SECTION file has five attributes Table 3. SECTION file Section no Course no Semester Year Instructor 1 08SCS11 1 2008 Dr.G.V.P 2 08SCS12 1 2008 Dr.G.V.P 3 08SCS13 1 2008 KNS 4 08SCS14 1 2008 Dr.S.R.K 5 08SCS151 1 2008 GRP 6 08SCS27 3 2008 Dr.G.V.P 6. Ex: GRADE_REPORT file has three attributes Table 4. GRADE_REPORT Register no Section no Grade 001 1 A 004 2 B 001 2 C CHARACTERISTICS OF DATABASE APPROACH 1. Single Repository of Data(SRD) a. A place where the data is stored b. A place where multiple databases or files are located for distribution

VIJAYA COLLEGE

Page 2

Database Management Systems

2. Self-describing a. Catalog which contains information such as structure of the file, b. Storage format and constraints 3. Multiple views of same data a. It allows multiple users to access the data database at same time. 4. Multi-user transaction processing a. It allows concurrency control s/w to ensure that several users trying to update the same data b. Ex: A seat reservation on a flight. ACTORS ON THE SCREEN 1. Maintenance of Dbase (single person) a. We need multiple people to design, development, and maintenance of a dbase 2. Database Administrator (BDA) a. Creates a database b. Managing the resources c. Creation of users accounts d. Providing security and authentication e. System recovery if system fails f. Manages the system poor response time 3. Database designer a. Indentifying the data to be stored in the database b. Planning the database design that meets all users requirements 4. End users a. Casual users(C users) i. There are people who use the Dbase occasionally b. Nave users (N users) i. There are people who constantly querying and updating the dbase without having much knowledge abt dbase ii. Ex: In railway, reservation clerks. c. Sophisticated users(SO Users) i. There are people who use the dbase for complex requirements ii. Ex: Engineers, Scientists, business analysts d. Standalone users(ST users) i. There are people who maintenance the dbase for personal use ii. Ex: http://dr.g.varaprasad.googlepages.com System analyst 1. Determine users requirements (N users) 2. Application programmers (S/w engineer) 3. Implement specifications as programs then they will test, debug, document, maintain Workers behind the Scene 1. There are many people who are involved in D&D&O of Dbase 2. DBMS system designers and implementers a. There are persons who design and implement DBMS modules 3. Tool developers a. There are persons who work for S/W tools to enhance the performance of a DBMS package 4. Operator and maintenance personnel a. There are people who are responsible for the actual running and maintenance of h/w and s/w environments Advantages of using DBMS 1. Controlling redundancy a. Duplication effort b. Wastage of storage space

VIJAYA COLLEGE

Page 3

Database Management Systems

c. Inconsistency in data 2. Restricting unauthorized access a. Some users will not be authorized to access all information in Dbase (Read the data but not update the dbase) b. Ex: Financial data is often considered confidential 3. Good UI a. DBMS should provide a menu driven S/W so that user can access the data without remembering(commands) 4. Providing multiple user interfaces a. Provides a variety of users interfaces 5. System should support various types of users with varying knowledge a. Query language interface for casual users b. Programming language interface for application programmers c. Formal and command interfaces for particular users d. Menu-driven interface and natural language interface for standalone users 6. Representing complex relationship among data a. Ex: Student name is Prasad in student table 7. Enforcing Integrity Constraints a. Each record in a table should have a proper semantic relationship with a record of another table. b. If employee works in dept no.5, then there must be dept table in which has dept no.5. 8. Providing backup and recovery a. It should provide facilities for recovering data from s/w and h/w failures A Brief History of Database Applications 1. We can use Dbase at different places as follows a. Railway/Flight reservation systems b. Universities c. Banking systems (Credit card systems) d. Online purchasing systems NOT TO USE A DBMS 1. There are few disadvantages of using DBMS as follows a. High level investment in h/w and s/w and training b. Too complex to use and maintain as this is a large piece of s/w c. Overhead for providing security, concurrency control, recovery and integrity functions 2. Dont use DBMS a. The application is simple b. Not going to change often c. Multiple-user access the data is not required d. Stringent Real time requirements for some programs Instance 1. Collection of information in dbase at a particular movement is called instance of dbase 2. Dbase changes over time as information is inserted or deleted 3. Ex: STUDENT relation Table 5. STUDENT relation Student name Register no Semester Major Varaprasad 001 1 CSE Prasad 004 2 ECE Schema

VIJAYA COLLEGE

Page 4

Database Management Systems

1. Overall design of dbase is called schema 2. Ex: Student database 3. STUDENT relation Table 6. STUDENT relation Student name Register no Semester Major Varaprasad 001 1 CSE Prasad 004 2 ECE Ex: COURSE relation Table 7. COURSE relation Course name Course no Credit Dept OS 08SCS11 3 CSE DS&A 08SCS12 4 CSE DBMS 08SCS13 4 CSE CSPA 08SCS14 4 CSE TFCS 08SCS151 3 CSE Ex: SECTION file Table 8. SECTION file Section no Course no Semester Year Instructor 1 08SCS11 1 2008 Dr.G.V.P 2 08SCS12 1 2008 Dr.G.V.P 3 08SCS13 1 2008 KNS 4 08SCS14 1 2008 Dr.S.R.K 5 08SCS151 1 2008 GRP 6 08SCS27 3 2008 Dr.G.V.P Ex: GRADE_REPORT file Table 9. SECTION file Register no Section no Grade 001 1 A 004 2 B 001 2 C Data base state: 1. Empty state 2. Initial state 3. current state 4. Final state Database System Utilities It helps DBA to mange the dbase 1. Loading utility 1. is used to load the existing data file into a dbase 2. Ex: Text file 2. Backup utility 1. Creates a backup copy of dbase 2. if failure occurs, we can use backup 3. Make more reliability 3. File reorganization utility 1. We use file reorganization to store the data 4. Performance utility 1. Monitors the dbase usage & provides static information to DBA Classification of Dbase Management System 1. First, we differentiate DBMS based on the data models 2. Presently, there are two types of database models used in commercial dbase a. Relational data model b. Object data model

VIJAYA COLLEGE

Page 5

Database Management Systems

3. Secondly, classification of DBMS is based on number of users supported by the system a. Single user system supports for single user b. Multi-user system supports for multiple users 4. Thirdly, number of sites over which dbase in distributed 5. Fourth, cost of DBMS

CHAPTER 2: ENTITY-RELATIONSHIP MODEL

ENTITIES AND ATTRIBUTES Attributes 1. Represents the properties of an entity 2. Ex: SSN, DOB, Name, Sal, etc. Entity 1. It is anything that exists in real world 2. Have multiple attributes 3. EX: Employee relation Figure 1. Employee relation. Entity set 1. Collection of similar entities with same properties 2. Ex: COMPANY, EMPLOYEE, TREE

Attributes types Single attribute or atomic attribute 1. Cant be subdivided further 2. Ex: in employee relation, SSN, Age, sex, sal Composite attribute(C attribute) 1. Can be subdivided further 2. Ex: in Employee relation, address Figure 2. Address attribute Single value attribute 1. Has single value for a particular entity set 2. Ex: In Employees relation, Age, Dept no Multi-value attribute 1. Can be divisible 2. Ex: Degree(B.E, B.Tech, M.E, M.Tech, PhD, DSc) Address H.NO Cros Main Street Area

City State County Pin code

EMPLOYEE SSN DOB Name Sal FN Min LN Add 3. Ex: Color(Red, Green, Yellow) Stored attribute 1. The value of certain attributes can not be obtained/derived from other attributes 2. Ex: In Employees relation, SSN is stored attribute 3. Ex: In student relation, Reg no is stored attribute Derived attribute 1. The value of attribute can be derived from other attributes 2. Ex: In Employees relation, age Key 1. A key is a minimal set of attributes of an entity set, which would uniquely identify an entity in an entity set. 2. Ex: In Employee relation, SSN is primary key Figure 3. Employee relation 3. Ex: Student Figure 4. Student relation. Domain of attributes

VIJAYA COLLEGE

Page 6

Database Management Systems

1. Each attribute is associated with set of values 2. Ex: Name : - the domain is [a-z], [A-Z] 3. Ex: number: Domain is [1-17] used in register number for M.Tech(CSE) E-R diagram 1. Strong entity set 1. Ex: Employee 2. Figure 5. Strong entity. 2. Weak entity set 1. Ex: Dependents 2. Figure 6. Weak entity. EMPLOYEE SSN DOB Name Sal FN MIN LN Add

STUDENT
Name Sem Major RegN
Employee Dependents

Attribute 1. Ex: RegN 2. 4. Multi-valued attribute 1. Ex: Degree 2. 5. Derived attribute 1. Ex: Age 2. Primary key 1. Ex: SSN 2. Relationship 1. Ex: Deposit 2. Participation 1. Partial Participation 2. _________ 3. Total portion 4. Relationship 1. It is an association among several entities 2. A relationship set is a mathematical relational among entities n=2 1. WorksFor 2. Ex: WORKSFOR

Age RegN SSN EMPLOYEE SSN Name Add DEPARTM ENT Dno DNam DAdd Dadd
3. Manages

VIJAYA COLLEGE

Page 7

Database Management Systems

4. Ex: MANAGES 5. If the no.of participating entity sets is two, then we called as binary relationship 6. Ex: Binary relationship MANAGES 7. If the no.of participating entity sets is three, then we called as ternary relationship(TR) 8. Ex: SUPPLY 9. Ex: USES 10. Ex: MANAGES

PROJECTS PNam EMPLOYEE SSN Name Add Date Name DEPARTMENT Dno DAdd Dadd PROJECT PN PNam EMPLOYEE SSN Name Add MACHINE S Mno MName Mcost SUPPLIER SSN Name Add PARTS PN PNam Pno Plocat EMPLOYEE DEPARTMENT EMPLOYEE DEPARTMENT
11. ER-Diagram for employees database Supervision WorkFor Supervisor Supervisee Manages Control Dependence of Works_on RELATIONSHIP 1:1 Relation Ex: one person has one passport 1:N Relation Ex: one student registers many subjects N:1 Relation

VIJAYA COLLEGE

Page 8

Database Management Systems

Ex: Many students have joined with a College N:N Relation Ex: Many employees works for many projects TOTAL PARTICIPATION Total Participation of Dept in Manges(Relation) MANAGES

EMPLOYE E Dept FN MM LN SSN Name Add Project Pno PNam Ploct Hours Startdate Dno DNam DAdd Dadd Name Sex Relationship DOB Dependent EMPLOYEE DEPARTMENT
RECURSIVE RELATIONSHIP SET (RR SET) 1. Where there is a relationship set which connects the same entity set is called RR set supervisor Supervision supervisee HoD supervisees many employees WEAK ENTITY SET 1. Entity set 2. Doesnt have a primary 3. Ex:
LOAN PAYMENT

REFINING THE ER-DRAM FOR COMPANY DBASE 1. Identify the entities 2. Identify the strong entities and weak entities a. Employees Strong entity b. Department Weak entity c. Project Strong entity d. Dependent Weak entity 3. Identify the relevant attributes a. Employee, SSN Name, add, sex, sal b. Department, Dno, dname, Dlocation c. Project , Pno, Pname, Plocation d. Dependent, Dename, DoB, Sex, Relationship 4. Identify the relationship sets a. Employee , Department, WorkFor b. Employee , Department, manages c. Department, Project, Controls d. Project, Employee, WorkOn e. Employee , Department, Dependent_Off f. Employee, Employee, supervision 5. Identify the cardinality ratio & participation constraints a. Works_For N 1

VIJAYA COLLEGE

Page 9

Database Management Systems

b. Works_On N N c. Manages 1 N d. Dependent_Off 1 N e. Controls 1 N f. Supervision

EMPLOYEE Pno LOAN Lno Type Amou Pdate P-amount PAYMENT

6. ER-Diagram for employees database Supervision WORKFOR Supervisor Supervisee MANAGES CONTROL DEPENDENCE_OF WORKS_ON 7. Write ER diagram of a company that keeps track of company and employee phone WORKS_IN Contains Has

EMPLOYEE FN MM LN DEPARTMENT Dno DNam SSN Name Add PROJECT Pno PNam Ploct Name Sex Hours Startdate DAdd Dadd Relationship DOB DEPENDENT SSN Name Add EMPLOYEE DNam DEPARTMENT Dno DAdd Dadd PN Model Add PHONE

VIJAYA COLLEGE

Page 10

Database Management Systems

8. Write ER diagram for a database that keeps track of text books used in course TEACHES REFEREES USES

SSN Name Add INSTRUCTOR CN Name COURSE BN Name Author TEXT BOOK Edition CHAPER 3. RELATIONAL MODEL AND RELATIONAL ALGEBRA
RELATIONAL MODEL 1. Relational model represents the dbase as collection of records 2. Ex: University dbase 3. Student table Student name Register no addr Phone DoB Sem Perc Varaprasad 001 N.S.Halli 5567895 22/9/79 1 70 Prasad 004 DS. Halli 5567897 10/6/65 2 60 4. Student table is called Student relation 5. In student relation, row is called a tuple 6. In student relation, column header is called an attribute DOMAIN, ATTRIBUTES, TUPLES, AND RELATIONS 1. A domain D is a set of atomic values a. Ex: in student relation b. Reg no: 10 alphanumeric characters i. Ex: 1BM99CS015 c. Name : characters i. Ex: Varaprasad d. Address i. alphanumeric characters ii. Ex: 456, N.S. Halli, Bangalore-19 e. Phone i. 7 digits ii. 5567895 f. DOB i. Date ii. Ex: 22/9/79 g. Sem i. 1 digit ii. Ex: 4 h. Perc i. Float{0-100} ii. Ex: 70 2. A Relation schema R, denoted by R(A1, A2,.An) is made up of a relation name R and a list of attributes (A1, A2,.An). 3. Degree of relation is number of attributes in a R a. Ex: in student relation(R), degree is 7 CHARACTERISTIC OF REMATISTICS OF RELATIONS 1. Ordering of tuples in a Relation(R)

VIJAYA COLLEGE

Page 11

Database Management Systems

a. R is defined as a set of tuples b. Tuples in R do not have any particular order c. In file, records are physically stored on disk so there is an order among the records d. Many logical models can bed used to sort records in R using SSN, name, age 2. Ordering of values within a tuple a. At logical level, the order of attributes, their values are not important b. Ex: A tuple can be considered as set of (<attribute>, <value>) pair 3. Values in the tuples a. Each value in a tuple is an atomic value. b. Multi-value attributes must be represented by separated relation c. Composite attributes are represented only their simple component attributes d. Null i. In employee relation, some employees do not have phone RELATIONAL CONSTRAINTS AND RELATIONAL DBASE SCHEMAS 1. Domain constraints a. Specifies the value of each attribute b. That must be atomic value from domain in R c. Ex: in student relation, name i. Name domain: domain is [a-z], [A-Z] ii. Reg no domain: domain is [1-17] 2. Key constraints and constraints on null a. A relation R is defined as a collection of tuples i. Ex: in employee relation, Varaprasad, SRK, Indramma, Prasad b. All records are distinct c. All records must have a key to identify the records uniquely i. Ex: in employee relation, SSN d. Super key i. Specifies uniqueness that no two distinct tuples in a state r of R can have same value ii. Has redundant attributes iii. In student relation, {Reg no, name, age} is a super key but it is not key of student e. Candidate key i. In R, it cant have same two tuples with same value ii. In some case, subset of set of attributes of relation has uniqueness property iii. Ex: in VEHICLE relation, Lice No and Engine No are candidate keys Lice No Engine No Make Model Year Ka-04EM7244 A68967 Hond Splendor + 2007 Ka-03M7244 B67856 Tata Nono 2008 iv. Lice no is considered as a primary key f. Primary key i. Specifies uniqueness that no two distinct tuples in a state r of R ii. EX: Student relation, Register no Student name Register no addr Phone DoB Sem Perc Varaprasad 001 N.S.Halli 5567895 22/9/79 1 70

VIJAYA COLLEGE

Page 12

Database Management Systems

Prasad 004 DS. Halli 5567897 10/6/65 2 60 iii. It should not have null value 1. Ex: Reg no : null RELATIONAL DATABASES AND RELATIONAL DATABASE SCHEMAS 1. So far we have discussed single relation and single schema 2. A relation dbase schema S has many relations with tuples 3. Represents as S={R1, R2, R3,.Rn} a. Ex: company database b. EMPLOYEE relation FN MN LN SSN DoB Add Sex Sal SuperSSN DN c. DEPARTMENT relation DN DName MGRSSN MGRSTDATE d. DEPT-LOCATION relation DN D-Location e. PROJECT relation PN PName P-Location DN f. WORKS_ON SSN PN Hours g. DEPENDENT ESSN Dependent_Name Sex BOD Relation 4. Foreign key a. It is a set of attributes of a relation, say, R2 whose values are required to mach values of some primary key of some other relation R1. b. Need to maintain contain the data consistency c. In Employee relation, SuperSSN, DN FN MN LN SSN DoB Add Sex Sal SuperSSN DN 5. Referential integrity a. A tuple in one relation should reference to an existing tuple in another relation b. Is used to maintain consistency among tuples of two relations c. Ex:Company database with pk and fk Relation Primary key Foreign key Referencing relation SSN SuperSSN Employee(SSN) EMPLOYEE DNo Department (DN) DEPARTMENT DN MGRSSN Employee(SSN) DN DN Department (DN) DEPTLOCATION D-Location PROJECT PN DN Department (DN) ESSN ESSN Employee(SSN) WORKS_ON PN PN Project (PN) DEPENDENT ESSN ESSN Employee(SSN) Dependent_Name d. EMPLOYEE relation FN MN LN SSN DoB Add Sex Sal SuperSSN DN e. DEPARTMENT relation DN DName MGRSSN MGRSTDATE f. DEPT-LOCATION relation DN D-Location g. PROJECT relation PN PName P-Location DN h. WORKS_ON SSN PN Hours i. DEPENDENT ESSN Dependent_Name Sex BOD Relation UPDATE OPERATIONS AND DEALING WITH CONSTRAINTS VIOLATIONS

VIJAYA COLLEGE

Page 13

Database Management Systems

1. Operations of the relational model can be categorized into retrievals and updates. 2. Three basic operations in RM a. Insert i. Is used to insert new record into a relation ii. We want insert new employee in a relation. FN MN LN SSN DoB Add Sex Sal SuperSSN DN iii. Ex: We want insert tuple whose name is Jerge W. Buss in employee relation iv. insert<Jerge , W, Buss, 766669999, 1960-05-06, 53 DS Lane, NS, M, 20000, 766669969, 5> ----- is possible FN MN LN SSN DoB Add Sex Sal SuperSSN DN Jerge W Buss 766669999 196005-06 53 DS Lane, NS M 20000 766669969 5 v. Ex: We want insert tuple whose name is Krishma R. Murthy in employee relation vi. insert<Krishma , R, Murthy, 766669998, 1940-04-03, 23 DS Lane, LS, M, 40000, 766669969, 5>----- is possible FN MN LN SSN DoB Add Sex Sal SuperSSN DN Jerge W Buss 766669999 196005-06 53 DS Lane, NS M 20000 766669969 5 Krishma R Murthy 766669998 194004-03 23 DS Lane, LS M 40000 766669969 5 vii. Ex: We want insert tuple whose name is Jerge J. Buss with SSN is null in employee relation viii. Ex: insert<Jerge , J, Buss, Null, 1980-07-06, 53 DS Lane, NS, M, 20000, 766669969, 5>--------- is not possible SSN is null ix. Ex: We want insert tuple whose name is Jerge J. Buss in DN 100000 in employee relation x. insert<Jerge , J, Buss, 6767669998, 1980-07-06, 53 DS Lane, NS, M, 20000, 766669969, 100000>---------- not possible DN 100000 is not there in dept relation xi. Ex: insert<Senior, W, Buss, 766669999, 1944-04-05, 43 NS Lane, LS, M, 54000, 766669969, 5>----- is not possible( previous record had same SSN) xii. Insert new record in WORKS_ON relation, insert <766669999, 5, 6>---- is possible SSN PN Hours 766669999 5 6 xiii. Insert new record in WORKS_ON relation, insert <766669998, 5, 7>---- is possible SSN PN Hours 766669999 5 6 766669998 5 7 b. Delete i. Is used to delete record a record from relation ii. Ex: Delete the WORKS_ON tuple with SSN= 766669999, and PN=5 (possible since this record is available in dbase)

VIJAYA COLLEGE

Page 14

Database Management Systems

SSN PN Hours 766669998 iii. Ex: Delete the EMPLOYEE tuple with SSN= 766669998 is not possible (since WORKS_ON refer to this tuple) FN MN LN SSN DoB Add Sex Sal SuperSSN DN 766669999 c. Update i. Is used to modify the values of some attributes in a relation ii. Ex: Update the sal of the EMPLOYEE tuple with SSN=766669999 to 32000 ------(Accepted) FN M N LN SSN DoB Add Se x Sal SuperSSN D N Jerge W Buss 7666699 99 196 00506 53 DS Lan e, NS M 3200 0 7666699 69 5 Krish ma R Murt hy 76666999 8 1940 -0403 23 DS Lane , LS M 4000 0 76666996 9 5 iii. Update the DN of the EMPLOYEE tuple with SSN=766669999 to 1 ------(Accepted) FN M N LN SSN DoB Add Se x Sal SuperSSN D N Jerg

VIJAYA COLLEGE

Page 15

Database Management Systems

e W Bus s 7666699 99 1960 -0506 53 DS Lan e, NS M 3200 0 7666699 69 1 iv. Update the SSN of the EMPLOYEE tuple with SSN=766669998 to 766669999 ------(not accepted, since 766669999 already there in dbase) BASIC RELATIONAL ALGEBRA OPERATIONS 1. Basic set of relational model operations constitute the relational algebra 2. Enables the user to retrieval requests 3. Set operations are to group a. First i. Select ii. Project iii. Join b. Second i. Union ii. Intersection iii. Set Difference iv. Cartesian product 4. Select operation a. Used to select a subset of the tuples from a relation that satisfy a condition b. Syntax is c. Ex: Select the tuples(employees) who work in dept no 4, or those whose salary grater than Rs.30,000 P.M

dno=4 ii. sal>30000 d. (dno=4 and sal>30000)

i.
(EMPLOYEE) (EMPLOYEE)

(EMPLOYEE)

e. Ex: Select the tuples who either work in dept no 4 or make over Rs.25000 P.Mor, those whos salary greater than 30,000 P.M, or work in dept no 5.

dno=4 ii. sal>25000 iii. dno=4 iv. sal>30000

i.

(EMPLOYEE)

(EMPLOYEE)

(EMPLOYEE)

(EMPLOYEE)

VIJAYA COLLEGE

Page 16

Database Management Systems

v. (dno=4 and sal>25000) OR (dno=5 and sal>30000)(EMPLOYEE) f. Boolean conditions i. AND 1. (cond 1 AND cond 2) is true if both cond 1 AND cond 2 are true, otherwise, it is false ii. OR 1. (cond 1 OR cond 2) is true if either cond 1 or cond 2 or both are true, otherwise, it is false iii. NOT 1. (NOT cond) is true if cond is false, otherwise, it is true 5. Project operation a. Used to select a subset of the attributes of a relation by specifying the names of the required attributes. b. Syntax is <Attribute list>(R) c. For example to get a list of all employees first name and last name and sal from EMPLOYEE relation d. Lname, Fname, sal(EMPLOYEE) 6. Sequence of Operations and the RENAME Operation a. We can use SELECT and PROJECT operations b. Find those customer who live in Bangalore c. Customer-name( Customer-city =Bangalore) (Customer) d. Ex: Retrieve the first name, last name and sal of employees who work in department 5, e. Lname, Fname, sal( (dno=5(EMPLOYEE)) f. We can write this query in another way i. DEP5-EMP

(dno=5(EMPLOYEE)

(EMPLOYEE)

ii. RESULTS Lname, Fname, sal(DEP5-EMP) g. We can write this query in another way i. TEMP

dno=5

ii. F(FIRSTNAME, LASTNAME, SAL) Lname, Fname, sal(TEMP) 7. Set Theoretical Operations a. Consider two relations R and S. b. UNION of R and S i. The union of two relations is a relation that includes all the tuples that are either in R or in S or in both R and S. ii. Duplicate tuples are eliminated. iii. Ex: Retrieve the SSNs of all employees who either work in dept 5 or directly supervise an employee who works in dept 5 iv. In this section, we can query using UNION operation v. DEP5-EMP vi. RESULT1

(dno=5(EMPLOYEE)
(DEP5-EMP) (DEP5-EMP)

SSN vii. RESULT2 SUPERSSN

viii. RESULT RESULT1 U RESULT2 c. Student relation Fist name Last name GV P RV L MV S HT L Instructor Fist name Last name LD R RV L

VIJAYA COLLEGE

Page 17

Database Management Systems

HV T HT L Student U Instructor Fist name Last name GV P RV L MV S HT L LD R HV T d. INTERSECTION of R and S i. The intersection of R and S is a relation that includes all tuples that are both in R and S. ii. Student Instructor Fist name Last name RV L HT L iii. Find all customers who have both loan and an accont iv. Customer-name(BORROWER) Customer-name(DEPOSIT) e. DIFFERENCE of R and S i. The difference of R and S is the relation that contains all the tuples that are in R but that are not in S. ii. Ex: R-S or S-R iii. Student Instructor Fist name Last name GV P MV S iv. Instructor - Student Fist name Last name LD R HV T 8. Cartesian Product a. Is binary operator which combines both relations b. Denoted by X c. Works on two sets d. It is sometimes called CROSS JOIN. e. S(A1, A2, .Am) X R(B1, B2, Bn)= m+n attributes DEPARTMENT Relation DNo Dame MGRSSN 1 CSE 40 2 ISE 56 3 MCA 44 PROJECT Relation PName Plocation DS Bangalore DBMS Chennai f. PROJECT Relation X DEPARTMENT Relation DNo Dame MGRSSN PName Plocation 1 CSE 40 DS Bangalore 1 CSE 40 DBMS Chennai 2 ISE 56 DS Bangalore 2 ISE 56 DBMS Chennai 3 MCA 44 DS Bangalore 3 MCA 44 DBMS Chennai g. We want to retrieve for each female employee a list of the names of her dependents i. FEMALE_EMP ii. EMPNAMES

sex=F

(EMPLOYEE)

Lname, Fname, sal

(FEMALE_EMP )

VIJAYA COLLEGE

Page 18

Database Management Systems

iii. EMP_DEPENDENTS EMPNAMES X DEPENDENTS iv. ACTUAL_DEPENDENTS

SSN=ESSN

(EMP_DEPENDENTS)

v. RESULT Lname, Fname, Dependent_name(ACTUAL_DEPENDENTS) 9. Join Operation a. Is used to combine the related tuples from two relations into single tuples b. Do cross product of the two relations( improved version of CP). c. Becomes more complex, tuples are removed within the cross product to make the result of the join more meaningful. d. Syntax is X e. Join operations are i. Natural Join ii. Theta join iii. Equijoin join iv. Outer Join f. Natural Join i. When we omit the condition during joining we call it is natural joining ii. Let S and R are two relation, they have attribute like S(A1, A2, .Am, C1, C2.Ck) and R(C1, C2.Cl, B1, B2, Bn) iii. C1, C2.Cl are common attributes iv. We will use simple of * v. Ex: Department DNo Dame 1 CSE 2 ISE 3 MCA PROJECT Relation PNo PName DNo 10 DS 2 20 DBMS 2 30 TFCS 1 40 OS 3 vi. Natural join of Department and PROJECT is PNo PName DNo Dame 10 DS 2 ISE 20 DBMS 2 ISE 30 TFCS 1 CSE 40 OS 3 MCA vii. Here, joining is done over the attribute of DNo of Department and DNo of PROJECT viii. In PROJECT Relation, DNo is FK g. Theta join i. Used joining of two relations ii. is similar like Cartesian Product iii. Here, we will use condition iv. Let R and S are relations v. If x is attribute of R and y is attribute of S, then RXS where xy vi. Where, is .,=,>,=, etc.. h. Equijoin join i. Let R and S are relations ii. Syntax is R X <joint_condtion> S iii. We want to retrieve the name of manager of each department. iv. DEPT_MGR DEPARTMENT X MGRSSN=SSN(EMPLOYEE) v. RESULT Dname, Fname, Lname(DEPT_MGR) i. Outer Join i. Is similar like CP and Equijoin join ii. Tuples of a relation dont participate in join is called dangling tuples iii. Outer join operations are

VIJAYA COLLEGE

Page 19

Database Management Systems

1. Left outer join 2. Right outer join 3. Full outer join iv. Left outer join 1. Similar to natural join 2. but keep all dangling tuples of R 3. Syntax is R X S STUDENT Relation Regno Name % Bracnh 1BMCS05 Ravi 89 CSE 1BMME02 Prasad 92 MEC 1BMIS15 Shankar 86 IS IBMCV24 Siva 98 CV AWARD Relation Regno Amount Year 1BMCS05 1000 2007 1BMME02 2000 92 BMCV24 1500 98 RESULT STUDENT X AWARD Regno Name % Bracnh Amount Year 1BMCS05 Ravi 89 CSE 1000 2007 1BMME02 Prasad 92 MEC 2000 92 1BMIS15 Shankar 86 IS Null Null IBMCV24 Siva 98 CV 1500 98 v. Right outer join 1. Similar to natural join 2. but keep all dangling tuples of S 3. Syntax is R X S vi. Full outer join 1. Similar to natural join 2. but keep all dangling tuples of both R and S 3. Syntax is R X S 10. Division Operation a. Is kind of query operation b. Ex: Retrieve the names of employees who work on all projects that john smith works_on c. SMITH

Fanme =John and Lname=Smith

WORKS_ON

(EMPLOYEE) SMITH)

PNo(WORKS_ON X ESSN=SSN e. SSN_PNOS ESSN, PNo( )

d. SMITH_PNO f. SSNS SSN_PNOS SMITH_PNO

g. RESULT Fname, Lname(SSNS*EMPLOYEE) 11. Grouping and Aggregation function a. All relational algebra operations are used to represent as relation consisting of rows and columns b. Aggregation functions are i. Sum, Age, Count, Maximum, Minimum ii. Group_attributes l Aggregate_functions(Relation) c. Retrieve dept no, no.of employees and their average salary i. RESULT(Dno, N, Avgsal) Dno lcount(SSN), Avg(sal)(EMPLOYEE) d. Participation the relation into groups e. Apply aggregate function to each group f. Retrieve dept no, no.of employees and their average age i. RESULT(Dno, N, Avgsal) Dno lcount(SSN), Avg(age)(EMPLOYEE) Examples of queries in Relational Algebra

VIJAYA COLLEGE

Page 20

Database Management Systems

g. Retrieve the name and address of all employee who work for CSE dept CSE_DEPT Dname=CSE(Dept) CSE_EMP (CSE_DEPT X Dno=Dno(EMPLOYEE) RESULT Fname, Lname(CSE_EMP) 12. For every project located in Stanford, list the projects numbers, the controlling, and dept managers, last name, address, birth of date. STANFFORD_PROJ Plocation= Stanford CONTROL_DEPT STANFFORD_PROJ X Dno=Dno(Dept) PROJ_DEPT_MGR CONTROL_DEPT X MGRSSN=SSN(EMPLOYEE) RESULT Pno, Dno, Fname, Lname, add, bod(PROJ_DEPT_MGR) 13. Find the name of employees who work on all projects controlled by dept 5.

Pno(Dno=5 b. EMP_PROJ(SSN, Pno) Essn, Pno

a. DEPT5_PROJ(Pno)

(PROJECT)

(Works_On)

c. RESULT_EMP_SSNS EMP_PROJ DEPT5_PROJ d. RESULT Fname, Lname(RESULT_EMP_SSNS* EMPLOYEE) 14. Make list of project numbers that involves an employee who last name is Smith, either as a worker or a manager of dept that controls the department.

ssn(Lname=Smith SMITH_WORKER_PROJ Pno

SMITH(ESSN) MGRS

(EMPLOYEE)

(WORKS_ON * SMITH)

15. List the name of managers who have atleat one dependent.

MGRSSN(Department)

//Find MGRSSNs

EMP_With_Deps(ssn), ESSN(Dependent) MGRS_With_Deps (MGRS EMP_With_Deps) Result Fname, Lname(MGRS_With_Deps *Employee) ER To Relational Mapping Rules (ER to Table) 1. ER diagram it doesnt mean that logical dbase design is over all(constraints and proper set of relations). 2. Strong entity to table Rule 1: a. For every strong entity E, creates a table R that includes all attributes like primary key, etc. b. Employee table SSN Name Sal Add DoB 3. Rule 2: Weak entity to table a. How to identify weak entity i. Doesnt have primary key ii. Has partial key iii. Participates as one-to-many relations iv. If the participating owner entity is deleted, then all its references must be deleted. b. Convert weak entity to table EMPLOYEE SSN DOB Name Sal Add c. Rule 3: Mapping rule for ER relationship sets. ii. Entity sets are related with binary relationship iii. Relationship sets are WorksFor, Manages, Dependents, WorksOn, Control, Supervision. 1. Employee Workfor Department 2. Employee Manages Department 3. Employee Dependentsof Dependents 4. Employee WorksOn Project

VIJAYA COLLEGE

Page 21

Database Management Systems

5. Department Control Project 6. Employee Supervision Employee iv. One-to-One cardinality 1. r is attribute in Employee full participation v. One-to-Many vi. Many-to-Many cardinality 1. Translating M-to-M relationship set to table 2. Creates a new table with relationship set name 3. A&B are relations, which contain primary keys of t and s to be composite P.K for new relation. a. Rule 4: Handling Multi-valued attribute 4. Each multi-valued attribute should be translated into table. a. Rule 5: Ternary relationship set to table 5. You want create a table with P.Ks of all three participating entities Correspondence between ER & Relation model ER model Relational model 1. Entity type 1. Entity relation 2. 1:1 or 1:N type 2. Foreign key
Dependents

dname Sex Address

3. M:N relationship type 3. Relationship relation and two F.Ks. 4. N-array relationship type 4. Relationship relation and n F.Ks. 5. Composite attribute 5. Set of simple component attributes 6. Value set 6. Domain 7. Key attribute 7. P.K or S.K Introduction to client/Server Architecture 1. No. of applications are executed in client/server environment 2. Server has more powerful than clients to keep the data 3. Client sends request to server using its IP address and port number 4. Based on request, the server response to the client 5. Advantages a. Centralized resource i. To avoid problem caused by redundant and inconsistent data b. Improve security i. It provides more security c. Server level administration i. As client doesnt play major role d. Scale network i. We can add new user or delete clients 6. Disadvantage a. Increased cost due to the technical complexity of the server b. Weak link(if server fails, network will be disconnected 7. Centralize Architecture a. Network management platform resides on a single computer i. Ex: E-mail, banking b. Used in commercial applications, network alerts and events c. Pros: i. Single location to view events and alerts, more security d. Cons: i. Single system is not redundant or fault tolerant ii. More queue if system has more number of users. iii. More accessing time if system has more number of users.

CHAPER 4. STRUCTURE QUERY LANGUAGE (SQL) Introduction about SQL 1. It is used to represent a relational database, 2. Table is used to represent for relation. a. Rows for tuples. i. Ex: Records

VIJAYA COLLEGE

Page 22

Database Management Systems

b. Columns for attributes. i. Ex: SSN, name, dept-no 3. SQL2 commands for data definition such as Create, Alter, and Drop. 4. CREATE command is used to create/specify a new relation. a. Ex: CREATE TABLE COMPANY EMMPLOYEE. 5. DROP command is used to remove schema in a Dbase. a. The options are CASCADE and RESTRICT. 6. The CASCADE operation is used to remove dbase schema and tables, domains, other elements. a. DROP SCHEMA COMPANY CASCADE. 7. The RESTRICT operation is used if it has no element in it. 8. ALTER table Company is used to change dbase table. 9. ALTER TABLE COMPANY DEPARTMENT ALTER MGRSSN SET DEFAULT. Data types and Domains 1. Numeric Ex: SSN. 2. Character Ex: String. 3. Date Ex: 2008/12/20 4. Time Ex: 10:30 AM Basic Queries in SQL Select<Attribute list>-> is also called mapping From<Table list> <Condition> Question 1: Retrieve the birth date and address of employee whose name is John B. Smith Select BoD, ADD From EMPLOYEE Where FN=JOHN and MN=B and LN=Smith Question 2: Retrieve the name and address of all the employees who work for the Research dept Select FN, LN, ADD, Form EMPLOYEE, DEPARTMENT Where DNAME =RESEARCH and DNUMBER= DNO Question 3: For ever product located in Stanford, list the project number, construing dept number, department manager, last name, address, DOB. Select PNO, DNO, LN, ADD, DoB From PROJECT, DEPARTMENT, EMPLOYEE Where DNUM=DNO add MGRSSN=ESSN and PLOCATION=STANDFORD Question 4: Retrieve the all employees who work in Dept 5. Select * From EMPLOYEE Where DNO=5.

VIJAYA COLLEGE

Page 23

Database Management Systems

Question 5: Retrieve the all employees who work in Research Dept. Select * From EMPLOYEE, DEPARTMENT Where DNAME=RESEARCH and DNO=DNUM Union : It is used to combine all records Question 6: Select FN, LN From STUDENT Union Select FN, LN From TEACHER. Intersect Question 7: Let us find out names of students as well as teachers Select FN, LN From STUDENT Intersect Select FN, LN From TEACHER. Minus Question 8: Let us find out the names of student who are not teacher Select FN, LN From STUDENT Minus Select FN, LS From TEACHER

Question 9: Retrieve all employees in Dep 5 whose salary is between $ 30,000 and $40,000. Select * From EMPLOYEE Where(Salary between 30000 and 40000) and DNO=5. Question 10: Retrieve all the employees whose address is in HOUSTON, TEXAS Select FN, LN From EMPLOYEE Where address like 1% HOUSTON, TEXAS% Question 11: Find all the employees who were born during the 1950s Select FN, LN From EMPLOYEE Where BOD like --5-,--,-Question 12: Retrieve the name of each employee who works in all projects Controlled by Dept 5

VIJAYA COLLEGE

Page 24

Database Management Systems

Select FN, LN From EMPLOYEE Where ((Select PNO From WORKS_ON Where SSN=ESSN) Contains(Select Pno from PROJECT where Pnum=5)) Question 13: Retrieve the names of employees who name no dependents Select FN, LN From EMPLOYEE Where not exists(Select * From DEPENDENT Where SSN=ESSN); Aggregate Functions 1. Count, Sum, Max, Min, Avg Question 14: Find the sum of the salaries of all the employees, the maximum salary, the minimum salary, the average salary Select Sum(Salary), Max(Salary), Min(Salary), Avg(Salary) from EMPLOYEE Question 15: Find the sum of the salaries of all employee of research department, as well as maximum salary, the minimum salary, the average salary in dept Select Sum(Salary), Max(Salary), Min(Salary), Avg(Salary) From EMPLOYEE, DEPARTMENT Where Dno=Dnum and Dname= Research Grouping and Aggregate Function 1. All relation algebra operations are used to represent a relation consisting of rows, columns. 2. It has Sum, Avg, Count, Maximum, Minimum functions 3.Group attributes 3 Aggregate functions Relation 4. Prediction the relation into groups 5.Apply aggregate function to each group Question 16: Retrieve Dept no, No of employees and their average salary Result (Dno, N, Avg Sal) Dno 3 Count(SSN), Avg(Sal) (EMPLOYEE) Question 17: Retrieve Depno, No of employee and their average age Result (Dno, N, Average) Dno 3 Count(SSN), Avg(Age) (EMPLOYEE) Examples of Queries in Relational Algebra. Question 18: Retrieve the name and address of all the employees who work for CSE dept CSE-Dept (Dname=CSE(DEPARTMENT) CSE-Emp (CSE-Dept Dno=Dno EMPLOYEE) Result FN, Add(CSE-Emp) Question 19: For every project located in Stanford, list the project number, the controlling and dept managers last name address, birth date Standford_Project (Plocation=Stafford(PROJECT)

VIJAYA COLLEGE

Page 25

Database Management Systems

Control_Dept (Stanford_Project Dno=Dno DEPARTMENT) Project_Dept_MGR (Control_Dept MGRSSM=SSN EMPLOYEE) Result (Pno, Dno, FN, LN, Add(Project_Dept_MGR) Question 20: Find the name of the employees who work on all projects controlled By dept no5 Dep5-Proj Pno(Dno=5(PROJECT)) //project number Emp-Proj ESSN, Pno(WORKS_ON) Result_Emp_SSNS Emp_Proj /Dept5_Proj Result FN, LN(Resul_Emp_SSNS) Question 21: Make a list of project numbers that involve an employee whose last name is smith, either as a worker or a manager of dept that controls the project: Smith (ESSN) SSN(LN=Smith) (EMPLOYEE) Insert command in SQL 1. Used to add single tuple into the dbase a. Ex: INSERT INTO EMPLOYEE VALUES (Richard, K, Kotti, 63332986, 1962-12-30, 98 Oak, Tack, TX, M, 37000, 986756999, 4); 2. We can also use to assign values for certain attributes for new tuple. a. INSERT INTO EMPLOYEE(Fname, Lname, Dno, SSN) VALUES (Varaprasad, Golla, 4, 98653257); Delete command in SQL 1. Used to remove tuple from a relation Question 22: Remove employee whose name is Brown Delete from EMPLOYEE Where LN=Brown Question 23: Remove employee whose SSN is 123456789 Delete from EMPLOYEE Where SSN=123456789 Question 23: Remove all employees who work in research department Delete from EMPLOYEE Where Dno in (Select Dnumber From Department Where Dname=Research) Update Command in SQL 1. Used to modify value of attribute 2. Ex:To change the location and controlling department number of project 10 to Bangalore and 5. a. Update Project Set Plocation=Bangalore and Dno=5; Where pnumber=10

VIJAYA COLLEGE

Page 26

Database Management Systems

View (Virtual table ) in SQL 1. It is a single table that is derived from other tables 2. Doesnt necessarily exist in physical form(virtual table) 3. Ex: Create View Works_On1 As Select Fn, Ln, Pname, Hours From Employee, Project, Works_On Where SSN=ESSN AND Pno=Pnumber 4. Create View Dept_Infor(Dept_name, N0_of_emps, Total_sal) As Select Danme, Count(*), Sum(Salary) From Department, Employee Where Dnumber=Dno Group by Dname; 5. Retrieve the last name, first name, of all employees who work on Project X Select FN, LN From Works_On1 Where Pname =Project View implementation and view update 1. Query modification, modifies view query into a query. 2. View materialization, involves physical creating a temporary view table 3. Ex: Update the Pname attribute of Jhon B Smith from Product X to Product Y. a. Update Works_On1 Set Pname=Product Y Where LN= Smith AND FN=Jhon AND Pname=Product Y; Specifying General Constraints as Ascertain 1. To specify the constraints that salary of an employee must not be grater than salary of the manager that employee works for Create Assertion Salary_Constraint Check (Not exists(Select * from Employee E, Employee M, Department D Where E.Sal>M.Sal AND E.Dno=D.Dnumber AND D.MGRSSN=M.SSN) Additional Features of SQL 1. Granting and revoking privileges to the users 2. Supports insert, delete, update, and select operations 3. SQL statements support general purpose languages(C, C++, Pascal) 4. has transaction control commands for concurrency control, recovery purpose Database program issues 1. Interface between application program and dbase 2. Records in dbase corresponds to records in a programming language should be distinct type. CHAPTER 5. DATABASE DESIGN 1 Need of normal formal forms 1. It is a logical design method, which minimizes data redundancy and reduces design flaws. 2. Consists of applying various normal forms to do database.

VIJAYA COLLEGE

Page 27

Database Management Systems

3. The normal forms break down large tables into smaller subsets. First Normal formal form (INF) Def: A relation(R) is said to be in 1NF if every attribute of R takes only single value(atomic value). 1. It contains no repeating attributes or groups of attributes. 2. No repeating columns within a row. 3. No multi-valued columns. 4. Ex: In employee relation, ESSN Name Age Sex Dependent 41 Dr. Varaprasad 28 M {Dr. S.R. Kirhna, Dr. Babu} 42 Dr. Ravi 32 M {Dr. Summa, Shankar} 5. It is not in 1NF. Since it dont have single value for attribute Dependent ESSN Name Age Sex Dependent 41 Dr.Varaprasad 28 M Dr. S.R. Kirhna, 41 Dr.Varaprasad 28 Dr. Babu 42 Dr.Ravi 32 M Dr. Summa 42 Dr.Ravi 32 M Shankar 6. Ex: In department relation Department no Department name MGRSSN Dept location 1 CSE 41 {Blore, Mysore} 2 ISE 42 {Blore, Pune} 7. It is not in 1NF. Since it dont have single value for attribute Dept location Department no Department name MGRSSN Dept location 1 CSE 41 Blore 1 CSE 41 Mysore 2 ISE 42 Blore 2 ISE 42 Pune Functional Dependency Def: In relation, X Y is full functional dependency(FD), if removal of attribute X, then it does not hold any more relation. ESSN Name // if removal of attribute ESSN doesnt hold a relation {ESSN, PNo} Hours// if removal of attribute ESSN, PNo doesnt hold a relation Partial Functional Dependency Def: In relation, if removal of any attribute, then it does hold still a relation. Ex:{ESSN, Name} Name, if you remove of attribute, Name, it still holds a relation Second Normal formal form (2NF) Def: A relation(R) is said to be in 2NF if ever non-prime attributes in R must be functionally dependent on the primary key(candidate key). EMPPROJECT relation -----------------------------------------------------------------------------------------------------------SSN PNo Hours Ename Pname Plocation -----------------------------------------------------------------------------------------------------------1. FD-1: SSN Ename. 2. FD-2: PNo {Pname, Plocation} 3. FD-3: {SSN, PNo} Hours

VIJAYA COLLEGE

Page 28

Database Management Systems

4. It satisfies 1NF but not in 2NF 5. if we remove PNo, it still hold a relation FD-1; 6. if we remove SSN, it still hold a relation FD-2; 7. Decompose EMPPROJECT relation into 3 parts a. S-E table, PPP table and SPH table 8. S-E table -----------------------SSN EName -----------------------9. PPP table ---------------------------------------------Pno Pname Plocation ---------------------------------------------10. SPH table --------------------------SSN Pno Hours --------------------------Third Normal Form(3NF) Def: A relation schema R is in 3NF, if it is in 2NF and no non-prime attribute of R is EMPDEPT relation ---------------------------------------------------------------------------------------------------------------SSN Name DoB Addr Dno Dname MGRSSN ----------------------------------------------------------------------------------------------------------------1. FD-1: SSN {Name, DoB, Addr, Dno} 2. FD-2: Dno {Dname, MGRSSN} 3. Here, Dno is non-prime attribute(violate 3NF) 4. .EMPDEPT relation is decomposed into employee and department. 5. Employee relation ------------------------------------------------------------------------SSN Name DoB Addr Dno ------------------------------------------------------------------------6. Department relation --------------------------------------------------Dno Dname MGRSSN --------------------------------------------------Boyce-Codd Normal Form(BCNF) Def: A Relation is in 3NF, if and only if every deterministic is a table is candidate key. Relation table -------------------------------------ABCD -------------------------------------1. FD has four attributes 2. Since FD-2 is violating BCNF 3. Spilt FD relation into R1 and R2 a. R1(A,C, D) with FD AC D; R2(C, B) with C B 4. R1 table ------------------------------

VIJAYA COLLEGE

Page 29

Database Management Systems

ACD -----------------------------5. R2 table ----------------CD ---------------CHAPTER 6. TRANSACTION MANAGEMENT
1. Transaction is a logical unit of dbase 2. Read operation : It read the data item in data base but cant not update the value in dbase T1 T2 Read_Item(X); X:=X-N; Read_Item(X); X:=X+N; 3. Write operation: It read the data item as well as update the item in data base. T1 T2 Read_Item(X); X:=X-N; Write_Item(X); Read_Item(X); X:=X+N; Write_Item(X); 4. Single user: Only one user can access the Dbase at time a. Ex: DoS is Single user OS. 5. Multi-user: Many users can access the Dbase currently a. Ex: Window XP is multi-user OS. 6. Multiprogramming : which allows computer to execute multiple programs 7. Interleaved: a. It allows system to execute one task for period of time then it suspends the task and executes other tasks. b. Reduces the cost c. Increases the delay d. Ex: We have one process and have to do three tasks Process A will do all three tasks T1 T2 T3 Read_Item(X); X:=X+N; Read_Item(X); X:=X+2N; Read_Itme(X) X:=X+3N Write_Item(X); Write_Item(X); Write_Item(X); 8. Parallel Processing: a. It has multiple processes to execute multiple tasks. b. Reduces the delay

T i m e T i m e

VIJAYA COLLEGE

Page 30

Database Management Systems

T i m e
c. Increases cost d. Ex: We have one process and have to three tasks e. Here, we have multiple processes (for example 3) Process A Process B Process C T1 T2 T3 Read_Item(X); Read_Item(X); Read_Itme(X) X:=X+N; X:=X+2N; X:=X+2N; Write_Item(X); Read_Item(X); Read_Itme(X) Why Concurrent Control Required 6. Reduces delay 7. Keeps CPU always busy with some jobs without waiting in idle. 8. Problems a. Last update problems b. A=25 T1 T2 Database Read_Item(A); // A=25 A:=A+100 Read_Item(A); // A=25 A:=AX2; Write_Item(A); //A=125 Read_item(B) Write_Item(A); Here, A has incorrect value because its update by T1 9. Temporary update problem a. A=25; b. B=50; T1 T2 Database Read_Item(A); // A=25 A:=A+100 Write_Item(A); // A=125 Read_Item(A); // A=125 A:=AX2; Write_Item(A); //A=250 Read_Item(B); //B=50 Fail //System has failed due to some problems and Dbase contains inconsistent information(In T1, some values are updated and some values are not updated(B)) Types of failures 1. Computer failures i. Due to H/W, S/W, N/W during a transaction 2. Transaction failures i. Due to integer over flow, division by zero 3. Local errors

T i m e T i m

VIJAYA COLLEGE

Page 31

Database Management Systems

e T i m e
i. You want to withdraw amount from the bank but you dont have sufficient amount in ur account. 4. Concurrent control problems i. Due to dead lock 5. Physical problems i. Power, Fire, Cyclone, Earthquake, Theft, etc Why Recovery is needed 1. Data should be recovered when computer fails due to transaction fails, Local errors, physical problems, concurrent control problems. Transaction states and additional operations 1. Begin state: We are about to start transaction 2. Read state: We are reading an item in a dbase Ex: Read_item(X) 3. Write state: We are writing item in database Ex: Write_item(X) 4. End_ Transaction state: We have finished all operations and are going to write all updates in Dbase. 5. Commit_ Transaction state: Successfully completed transaction 6. Rollback state: Transaction has failed due to some problems.

Read, write Begin Commit End transaction Abort Abort Desirable properties of transactions 1. Transaction posses several properties 2. Atomicity: A transaction is an atomic unit of processing. It is either performed a task or not. 3. Consistency Preservation: A transaction is consistency preserving if its complete of dbase from one state to another. 4. Isolation: A transaction is being executed in isolation from other transactions. T1 should not interface with T2 while execution of T1. Active Partial committed Committed Failed Terminated 5. Durability: The changes applied to the dbase by using committed operation must persist in a dbase. Schedules of transactions: 1. Schedule has n transactions T1, T2, T3..Tn. 2. Schedule has ordering operations subject to constraint.
T1 T2

VIJAYA COLLEGE

Page 32

Database Management Systems

Read_Item(X); X:=X+100 Read_Item(X); X:=X*2; Write_Item(X); Read_item(Y) Write_Item(X);

3. Sa: r1(x), r2(x), w1(x), r1(y), w2(x);

T1 T2 Read_Item(X); X:=X+100 Write_Item(X); Read_Item(X); X:=X*2; Write_Item(X); Read_item(Y) Fail

4. Sb : r1(x), w1(x), r2(x), w2(x), r1(y, a1 ; Serial Schedule : 1. Execute all operations of transaction T1 followed by all operations of transaction T2.
T1 T2 Read_Item(X); X:=X-N Write_Item(X); Read_Item(Y); Y:=Y+N; Write_Item(Y); Read_item(X) X:=X+M; Write_Item(X);

2. Execute all operations of transaction T2followed by all operations of transaction T1.

T1 T2 R Read_item(X) X:=X+M; Write_Item(X); Read_Item(X); X:=X-N Write_Item(X); Read_Item(Y); Y:=Y+N; Write_Item(Y);

T i m e T i m e T i m e T

VIJAYA COLLEGE

Page 33

Database Management Systems

i m e 3. We are executing transactions without interleaved operations Non Serial Schedule 4. We are executing transactions with interleaved operations
Read_Item(X); X:=X-N Read_item(X) X:=X+M; Write_Item(X); Read_Item(Y); Write_Item(X); Y:=Y+N; Write_Item(Y); Read_Item(X); X:=X-N Write_Item(X); Read_item(X) X:=X+M; Write_Item(X); Read_Item(Y); Y:=Y+N; Write_Item(Y);

Serializable A Schedule S of n transactions is serializable, if it is equivalent to some serial schedule of the same transaction. X=100 S1 S2
Read_Item(X); Read_Item(X); X:=X+10 X:=X*1.1 Write_Item(X); Write_Item(X);

X=110; X=110;

VIJAYA COLLEGE

Page 34