Introduction

Professo r I I T Madras

Introduction
What is a Database? A collection of related pieces of data: Representing/capturing the information about a real-world enterprise or part of an enterprise. Collected and maintained to serve specific data management needs of the enterprise. Activities of the enterprise are supported by the database and continually update the database.

Prof P Sreenivasa Kumar Department of CS&E, IITM

An Example
University Database: Data about students, faculty, courses, researchlaboratories, course registration/enrollment etc. Reflects the state of affairs of the academic aspects of the university. Purpos : To keep an accurate track of the academic e activities of the university.

Prof P Sreenivasa Kumar Department of CS&E, IITM

Database Management System (DBMS)

A general purpose software system enabling:

Creation of large disk-resident databases. Posing of data retrieval queries in a standard manner. Retrieval of query results efficiently. Concurrent use of the system by a large number of users in a consistent manner. Guaranteed availability of data irrespective of system failures.

Prof P Sreenivasa Kumar Department of CS&E, IITM

OS File System Storage Based Approach

Files of records used for data storage data redundancy wastage of space maintaining consistency becomes difficult Record structures hard coded into the programs structure modifications hard to perform Each different data access request (a query) performed by a separate program difficult to anticipate all such requests Creating the system requires a lot of effort Managing concurrent access and failure recovery are difficult

Prof P Sreenivasa Kumar Department of CS&E, IITM

DBMS Approach
DBM S separation of data and metadata flexibility of changing metadata program-data independence Data access language standardized SQL ad-hoc query formulation easy System development less effort required concentration on logical level design is enough components to organize data storage process queries, manage concurrent access, recovery from failures, manage access control are all available
Prof P Sreenivasa Kumar Department of CS&E, IITM 6

Data Model
Collection of conceptual tools to describe the database at a certain level of abstraction. Conceptual Data Modela high level description useful for requirements understanding. Representational Data Model describing the logical representation of data without giving details of physical representation. Physical Data Model description giving details about record formats, file structures etc.

Prof P Sreenivasa Kumar Department of CS&E, IITM

E/R (Entity/Relationship) Model

A conceptual level data model. Provides the concepts of entitie , relationship and attribute . s s The University Database s Context Entities: studen , faculty , cours , department etc. t member e s Relationships: enrollmen relationship between student & course, t employmen relationship between faculty t member, department etc. Attributes: nam , rollNumbe , addres etc., of studen entity, e r t nam , empNumbe , sphoneNumbe etc., of faculty e r r entity etc. More details will be given in the E/R Model Module.
Prof P Sreenivasa Kumar Department of CS&E, IITM 8

Representational Level Data Model

Relational Model : Provides the concept of a relation. In the context of university database:
Relation name Attributes studen t SName RollNumber JoiningYear BirthDate Program Dept Sriram CS04B123 2004 15Aug1982 BTech CS
Data tuple

Relation scheme: Attribute names of the relation. Relation data/instance: set of data tuples. More details will be given in Relational Data Model Module.
Prof P Sreenivasa Kumar Department of CS&E, IITM 9

Data versus Schema or Meta-Data

DBMS is generic in nature not tied to a single database capable of managing several databases at a time Data and schema are stored separately. In RDBMS context: Schema table names, attribute names with their data types for each table and constraints etc. Database definition setting up the skeleton structure Database Loading/populating storing data
Prof P Sreenivasa Kumar Department of CS&E, IITM 10

Abstraction Levels in a DBMS: Three-Schema Architecture

View Level(VL) scheme

V1

V2
VL

V3

Vm

Set of views

LL mapping

LDI

Logical Level(LL) scheme

R1
LL

R2

PL mapping

Rn

Set of relations

PDI

Physical Level(PL) Scheme

F1

F2

Fp

Data: set of files/index files

Prof P Sreenivasa Kumar Department of CS&E, IITM

11

Three-schema Architecture(1/2)
View Level Schema view describes an aspect of the database relevant to a Each particular group of users. For instance, in the context of a library database: Books Purchase Section Issue/Returns Management Section Users Management Section Each section views/uses a portion of the entire data. Views can be set up for each section of users.

Prof P Sreenivasa Kumar Department of CS&E, IITM

12

Three-schema Architecture(2/2)
Logical Level Schema Describes the logical structure of the entire database. No physical level details are given. Physical Level Schema Describes the physical structure of data in terms of record formats, file structures, indexes etc. Remark s Views are optional - Can be set up if the DB system is very large and if
easily identifiable user-groups exist

The logical scheme is essential Modern RDBMSs hide details of the physical layer
Prof P Sreenivasa Kumar Department of CS&E, IITM 13

Physical Data Independence

The ability to modify physical level schema without affecting the logical or view level schema. Performance tuning modification at physical level creating a new index etc. Physical Data Independence modification is localized achieved by suitably modifying PL-LL mapping. a very important feature of modern DBMS.
Three Schema Arch Three Schema Arch

Prof P Sreenivasa Kumar Department of CS&E, IITM

14

Logical Data Independence

The ability to change the logical level scheme without affecting the view level schemes or application programs Adding a new attribute to some relation no need to change the programs or views that dont require to use the new attribute Deleting an attribute no need to change the programs or views that use the remaining data view definitions in VL-LL mapping only need to be changed for views that use the deleted attribute
Three-schema Architecture

Prof P Sreenivasa Kumar Department of CS&E, IITM

15

Development Process of a Database System (1/2)

Step 1. Requirements collection Data model requirements various pieces of data to be stored and the interrelationships. presented using a conceptual data model such as E/R model. Functional requirements various operations that need to be performed as part of running the enterprise.
acquiring a new book, enrolling a new user, issuing a book to the user, recording the return of a book etc.

Prof P Sreenivasa Kumar Department of CS&E, IITM

16

Development process of a database system (2/2)

Step 2. Convert the data model into a representational level model typically relational data model. choose an RDBMS system and create the database. Step 3. Convert the functional requirements into application programs programs in a high-level language that use embedded SQL to interact with the database and carry out the required tasks.

Prof P Sreenivasa Kumar Department of CS&E, IITM

17

Architecture of an RDBMS system

GUI/parameter values Application programs Appln Pgm compiler Query compiler Compiled Appln pgms Trans Manager RDBMS Run Time System

Ad-hoc queries (Analyst) DDL Commands Control Commands (DBA )

Query optimizer

Buffer Manager

Meta data data

DDL and other command processor

Recovery Manager

Log

Disk Storage

Prof P Sreenivasa Kumar Department of CS&E, IITM

18

Architecture Details (1/3)

Disk Storage: Meta-data schema - table definition, view definitions, mappings Data relation instances, index structures statistics about data Log record of database update operations essential for failure recovery DDL and other command processor: Commands for relation scheme creation Constraints setting Commands for handling authorization and data access control

Prof P Sreenivasa Kumar Department of CS&E, IITM

19

Architecture Details (2/3)

Query compiler SQL adhoc queries Compiles update / delete commands Query optimizers Selects a near optimal plan for executing a query - relation properties and index structures are utilized Application Program Compiler Preprocess to separate embedded SQL commands Use host language compiler to compile rest of the program Integrate the compiled program with the libraries for SQL commands supplied by RDBMS
Prof P Sreenivasa Kumar Department of CS&E, IITM 20

Architecture Details (3/3)

RDBMS Run Time System: Executes Compiled queries, Compiled application programs Interacts with Transaction Manager, Buffer Manager Transaction Manager: Keeps track of start, end of each transaction Enforces concurrency control protocols Buffer Manager: Manages disk space Implements paging mechanism Recovery Manager: Takes control as restart after a failure Brings the system to a consistent state before it can be resumed

Prof P Sreenivasa Kumar Department of CS&E, IITM

21

Roles for people in an Info System Management (1/2)

Naive users / Data entry operators Use the GUI provided by an application program Feed-in the data and invoke an operation - e.g., person at the train reservation counter, person at library issue / return counter No deep knowledge of the IS required Application Programmers Embed SQL in a high-level language and develop programs to handle functional requirements of an IS Should thoroughly understand the logical schema or relevant views Meticulous testing of programs - necessary
Prof P Sreenivasa Kumar Department of CS&E, IITM 22

Roles for people in an Info System management (2/2)

Sophisticated user / data analyst: Uses SQL to generate answers for complex queries DBA (Database Administrator) logical scheme Designing the Creating the structure of the entire database Monitor usage and create necessary index structures to speed up query execution Grant / Revoke data access permissions to other users etc.

Prof P Sreenivasa Kumar Department of CS&E, IITM

23

Text Books
Ramez Elmasri and Shamkant B Navathe, Fundamentals of Database Systems , 3rd Edition, Addison Wesley, 2000. Raghu Ramakrishnan and Johannes Gehrke, Database Management Systems , 3rd Edition, McGraw Hill, 2003. A Silberschatz, H F Korth and S Sudarshan, Database System Concepts , 5th Edition, 2006. H Garcia-Molina, J D Ullman, and Jennifer Widom, Database Systems-The Complete Book, Pearson Education, 2004.
Prof P Sreenivasa Kumar Department of CS&E, IITM 24