Database Review

Outline

DefinitionsHistoryBasic concepts of DBMSData ModelsRelational databaseNormalization

Definitions

Database: an organized collection of dataRelational database: a database based on

the relational modelDatabase management system: a

computer program (or a suite of programs)designed to manage a database, and runoperations on the data requested bynumerous consumers

Source: www.en.wikipedia.org

History

1960's: Two main data models were developed:network model (CODASYL - COnference onDAta SYstems Languages) and hierarchical(IMS - Information Management System).

1970-72: Edgar Frank Codd proposed relationalmodel for databases - disconnects the schema(logical organization) of a database from thephysical storage methods.

History

1970's: Two main prototypes for relational systems weredeveloped during 1974-77. Ingres: Developed at UCB

led to Ingres Corp., Sybase, MS SQL Server, Britton-Lee,Wang's PACE.

used QUEL as query language.System R: Developed at IBM San Jose

led to IBM's SQL/DS & DB2, Oracle, HP's Allbase, Tandem'sNon-Stop SQL.

used SEQUEL (Sequential English QUEry Language) asquery language.

The term Relational Database Management System(RDBMS) is coined during this period.

History

1976: P. Chen proposed the Entity-Relationship(ER) model for database design.

Early 1980's: Commercialization of relationalsystems begins

Mid-1980's: SQL (Structured Query Language)becomes standardDB2 becomes IBM's flagship product.Development of the IBM PC gives rise to many DB

companies and products such as RIM, RBASE 5000,PARADOX, OS/2 Database Manager, Dbase III, IV(later Foxbase, even later Visual FoxPro), Watcom SQL.

History

Early 1990's: companies offering increasingly complexproducts at higher prices. development centers on client tools for application such as

PowerBuilder (Sybase), Oracle Developer, VB (Microsoft),etc.

Client-server model for computing becomes the norm forfuture business decisions.

Development of personal productivity tools such asExcel/Access (MS) and ODBC.

This also marks the beginning of Object DatabaseManagement Systems (ODBMS) prototypes.

Mid-1990's: Internet/WWW appears.Web/DB grows exponentially.

History

Late-1990's: Boom in Web/Internet/DBconnectors.Active Server Pages, Front Page, Java Servlets, JDBC,

Enterprise Java Beans, ColdFusion, Dream Weaver,Oracle Developer 2000, etc.

Open source solution come onlinewidespread use of gcc, cgi, Apache, MySQL, etc.Online Transaction processing (OLTP) and online

analytic processing (OLAP) comes of age with manymerchants using point-of-sale (POS) technology on adaily basis

History

Early 21st century: Decline of the Internetindustry as a wholesolid growth of DB applications continues.More interactive applications appear with use of PDAs,

POS transactions, consolidation of vendors, etc. Threemain (western) companies predominate in the large DBmarket: IBM (buys Informix), Microsoft, and Oracle.

Future trendsLarge data sizeMobile computing

Problems with Conventional FileSystemsData redundancy and inconsistencyDifficulty in accessing dataData isolationMultiple usersSecurity problems Integrity problems

DBMS

Characteristics:Self-containedInsulation between programs and dataData abstractionSupport of multiple views of the data

DBMS

Intended uses: Controlling redundancy Sharing of data - concurrency control Restricting unauthorized access Providing multiple interfaces Representing complex relationships among data Enforcing integrity constraints Providing backup and recovery

DBMS Advantages

Potential for enforcing standardsFlexibilityReduced application development timeAvailability of up-to-date informationEconomies of scale

Data Abstraction

Data abstraction is an abstract view of thedatabase system

Three levels of data abstractionPhysical level - how data are stored

physicallyConceptual level - what data are stored,

relationship among dataView level - which parts of data a specific

user can view

Data Models

Object-based logical data modelER modelObject-oriented model

Record-based logical data modelRelational modelNetwork modelHierarchical model

Physical data model

E-R Model

Consists of entities (objects) andrelationships among those entities

Each entity has a set of attributesdescribing it

Can be described using E-R diagram

E-R Diagram

Graphical expression of a logical structure ofa databaserectangles: represent entity sets.ellipses: represent attributes.diamonds: represent relationships among entity

sets.lines: link attributes to entity sets and entity sets

to relationships.

http://wofford.org/ecs/DataAndVisualization/ermodel/material.htm

Example of E-R Diagram

http://www.conceptdraw.com/products/img/ScreenShots/cd5/software/Chen_ERD.gif

Relational Model

Represents data and relationships withtables

Attributes are represented as columns ineach table

Each column must have a unique name

Example of Relational Database

Sirima Li3IT1

Santi Li4Math2

Anne Taylor1IT1Faculty_NameFaculty_IDDept_NameDept_ID

$2300Santi Li4$2000Sirima Li3$2500Anne Taylor1Faculty_SalaryFaculty_NameFaculty_ID

Physical Data Model

Describes data at the lowest levelDepends on physical DBMS

Technical Terms Comparison

FieldColumnAttributeRecordRowTupleEntityTableRelation

A.K.ACommonName

FormalName

Relational Operations

Query: selectUpdate: insert, update, deleteTables manipulation: create table, drop

table, alter table Index manipulation: Create index, drop

indexView manipulation: Create view, drop view

SELECT

Select a subset of tuples in a relationSELECT field(s) FROM table_nameWHERE condition

ORDER BY field(s)ExampleSELECT customer_name,credit_limit,addressFROM customer WHERE slsrep_number = 6ORDER BY credit_limit, current_balance

INSERT

Add a single tuple to a relationINSERT INTO table_name VALUESvalues to be inserted in columns

Example:INSERT INTO slsrep

VALUES (10, ‘Mark’, ‘1400 Joyce St.’,0.00, 0.05)

DELETE

Remove tuples from a relationDELETE table_name WHERE condition

ExampleDELETE customer WHERE name=‘Mark’

UPDATE

Modify attribute values of one or moreselected tuples

UPDATE table_name SET column = valueWHERE condition

ExampleUPDATE customer SET name = ‘Ann’WHERE customer_number = 100

Normalization

The process of converting a complex datastructure intosimplest structuremost stable structure

By removing redundant attributes, keys,and relationships from a conceptualdata model

1NF

First normal form (1NF) entity is achievedby removing repeating or multivaluedattributes to another, child entity.

Example

ORA

DTCASI

TOS

Cust

$6

$5$12

$10

$10$5

Prd_Price

Saw

PlierHammer

Hammer

HammerSaw

Prd_Dscr

202031006005

515

31

031006004151021006003

2010

12

021006001

Prd_QTY

Prd_ID

Order_Date

Order_ID

Example

First Normal Form

ORNDTCDTCASITOSTOS

Cust

10$5Saw2021006001

15$12Hammer1031006004$6

$5$10

$10

Prd_Price

Saw

PlierHammer

Hammer

Prd_Dscr

202031006005

53031006004151021006003

201021006001

Prd_QTY

Prd_ID

Order_Date

Order_ID

2NF

First normal form entities can be reducedto second normal form (2NF) by removingattributes that are not dependent on thewhole primary key.

Example

First Normal Form

ORNDTCDTCASITOSTOS

Cust

10$5Saw2021006001

15$12Hammer1031006004$6

$5$10

$10

Prd_Price

Saw

PlierHammer

Hammer

Prd_Dscr

202031006005

53031006004151021006003

201021006001

Prd_QTY

Prd_ID

Order_Date

Order_ID

Example

Second Normal Form

10$5Saw2021006001

15$12Hammer1031006004$6

$5$10

$10

Prd_Price

Saw

PlierHammer

Hammer

Prd_Dscr

202031006005

53031006004151021006003

201021006001

Prd_QTY

Prd_ID

Order_Date

Order_ID

ORA005DTC004ASI003TOS001

CustOrder_ID

3NF

Second normal form entities can bereduced to third normal form (3NF) byremoving attributes that depend on other,nonkey attributes (other than alternativekeys).

Example

Second Normal Form

10$5Saw2021006001

15$12Hammer1031006004$6

$5$10

$10

Prd_Price

Saw

PlierHammer

Hammer

Prd_Dscr

202031006005

53031006004151021006003

201021006001

Prd_QTY

Prd_ID

Order_Date

Order_ID

Example

Third Normal Form

102021006001

151031006004202031006005

53031006004151021006003

201021006001

Prd_QTY

Prd_IDOrder_DateOrder_ID

$5Plier3$5Saw2

Hammer

Prd_Dscr

$101

Prd_Price

Prd_ID

Denormalization

the process of attempting to optimize theperformance of a database by addingredundant data

sometimes necessary because currentDBMSs implement the relational modelpoorly

Source: www.en.wikipedia.org