Software Tools - Instructinstruct.uwo.ca/engin-sc/se352b/notes/HG-SE352Lect-W06... · 2006-03-04 · SE352b 3-Mar-06 (c) Hamada Ghenniwa, ECE, UWO 3 Feb 24, 2006 SE352b, ECE,UWO,

SE352b 3-Mar-06

(c) Hamada Ghenniwa, ECE, UWO 1

Feb 24, 2006 SE352b, ECE,UWO, Hamada Ghenniwa 1

SE352b Software Engineering SE352b Software Engineering Design ToolsDesign Tools

W6: W6: Database TechnologyDatabase Technology

Operational DatabaseOperational Database


Software ToolsSoftware Tools

Source Procedure A

Source Procedure B

Source Procedure C

Systems ProgrammingTools

Programming LanguageTools

DatabaseSystem

SE352b 3-Mar-06



DB Related IssuesDB Related Issues

Data Models

Schemas

Schema Architecture

DBMS Languages

DBMS Component Modules

Architecture of DBMS

Quality of Design in Relational DB

SQL in general context

Query Processing

Interacting with DBs


Classification of Classification of DBMSsDBMSs

Based on the data model used:

Traditional: Relational, Network, Hierarchical

Emerging: Object-oriented, Object-relational

Based on the number of users support

single-user (typically used with micro-computers)

multi-user (most DBMSs).

SE352b 3-Mar-06



Classification of Classification of DBMSsDBMSs(Cont(Cont.).)

Based on architecturecentralized

consists of a single DBMS managing a single DB

distributed consists of a single DBMS (calleddistributed DBMS, or DDBMS) that manages several DB that might be located at different computing systems

DBMS

DB

Centralized DBS

DDBMS

DB DB

Distributed DBS


Classification of Classification of DBMSsDBMSs(Cont(Cont.).)

A multi-database system (MDBS) supports operations on multiple DBSs,

each consists of separate DBMS or DDBMS

MDBMS

DB DB

DBMS DDBMS

Multi-DBS

DB

SE352b 3-Mar-06



33--Level SchemaLevel Schema

Internal/Local

External

Conceptual

MDBS: 5MDBS: 5--Level SchemaLevel Schema

To support the main characteristics of MDBS, a 5-level schema architecture is required

an extended version of the 3-level schema

1. External schema

2. Federated schema

an integration of multiple export schemas

3. Export schema

a subset of the component schema that is available to the FDBS

4. Component schema

the conceptual schema of a component

5. Local schema


DBMS

DB

Centralized DBS

MDBMS

DB DB

DBMS DDBMS

DB

MDBS: 5MDBS: 5--Level SchemaLevel Schema

DBDB--Component 1Component 1 DBDB--Component 2Component 2

MDBMS

DB DB

DBMS DDBMS

DB


Internal/Local

External

Conceptual


ExternalExternal External External

Federated

Export

Component1

Local

Federated Federated

Export

Component2

Local

Export

SE352b 3-Mar-06



ThreeThree--Schema ArchitectureSchema Architecture(Cont.)(Cont.)

DATABASE

End User End User

External LevelExternal LevelExternal Schema

VIEWExternal Schema

VIEWExternal Schema

VIEWExternal Schema

VIEW

Conceptual LevelConceptual Level Conceptual SchemaConceptual Schema

Internal LevelInternal Level Internal SchemaInternal Schema


Views Views Example…Example…

Section Id

YearSemesterGradeCourse Number

Student TranscriptStudent NameTRANSCRIPT

Database

Data generated on the flyUsing subject areas entitiesData generated on the flyUsing subject areas entities

Brown

Smith

13599FallACS3380

10299SpringBCS3320

9298FallACS1310

8598FallAMATH2410

11299FallBMATH2410

11999FallCCS1310

Database Definition

(MetaMeta--DataData)

DepartmentCreditCourse NumberCourse NameCourse NameCOURSE

InstructorYearYearSemesterSemesterCourse NumberSection IdentifierSection IdentifierSECTION

MajorClassStudent NumberNameNameSTUDENT

SE352b 3-Mar-06




Data Models

Schemas

Schema Architecture

DBMS Languages



Business Data Types



Query Processing



Business Data TypesBusiness Data Types

Operational Data

Informational Data DatabaseSystem

OperationalDatabase

Tools

InformationalDatabase

Tools

SE352b 3-Mar-06



Operational DataOperational DataThe data used to run the day-to-day business

Typically stored in a relational database

possibly, in legacy hierarchical or flat file formats as well

Some characteristicsUpdated often and through online transactions

Non-historical data (not more than three to six months old)

Optimized for transactional processing

Highly normalized in the relational database for efficient update, maintenance, and integrity


Informational DataInformational Data

Or information is a representation of the data in a way makes analysis much easier

Analysis can be in the form of decision support

executive information systems

report generation

more in-depth statistical analysis

Created from the wealth of existing operational data

SE352b 3-Mar-06



RelationalRelational Database DesignDatabase Design

∗ Conceptual design∗ ER Model is used at this stage to capture

the core business/application concepts and rules∗ which is then translated to a relational schema

• Schema refinement• Normalization: to tune the schema for efficiency

• check relational schema for redundancies and related anomalies

Physical database design tuning• Consider typical workloads and further refine the database design


Conceptual DesignConceptual Design

• What?• the entitiesentities and relationshipsrelationships in the enterprise

•• informationinformation about these entities and relationships should we store in the database

• the integrity constraintsintegrity constraints that hold

• Represent this information pictorially in ER diagramsER diagrams,

SE352b 3-Mar-06



Conceptual Design Conceptual Design ER Diagrams ER Diagrams ---- the basicsthe basics

Entitiesreal world “objects”,

e.g. Students, Courses, Routes, Climbers. Entity sets are drawn as rectangles:

Attributesthings typically used as column names,

e.g. Name, Age, Height.

Relationshipsrelationships between entities,

e.g. a student enrolls in a course, a climber climbs a route, etc. :


ER: ExampleER: Example

IDEF1X Notationhttp://www.idef.com/IDEF1X.html

Account ID

ACCOUNT

Account Type

Own

CLIENT

Client Name

Client IDClient ID

SE352b 3-Mar-06



Relational:Relational:DefinitionsDefinitions

Keya set of attributes of a relation whose values are unique within the relation

If there is more than one key, they are called Candidate KeysCandidate Keys

Prime attributePrime attributeone which is a member of any key

Nonprime attribute

one which is not prime

primary keyarbitrarily designated candidate keydesignated candidate key

all other candidate keys are Secondary KeysSecondary Keys


Client IDClient IDPrimary KeyNot Nullifyable

ER: ER: Identity & Functional DependencyIdentity & Functional Dependency

IDEF1X Notation

Account ID

ACCOUNT

Account Type

CLIENT

Account ID

ACCOUNT

Client NameAccount Type

Client IDOwn

Client ID (FK)

OwnClient ID (FK) Foreign Key

CLIENT

Nonprimeattributes

Client Name

Client ID

Not Nullifyable

Nullifyable

SE352b 3-Mar-06



Associative EntitiesAssociative Entities

An associative entity is a relation that contains data, and so may be treated as an entity.

Client ID

CLIENT

Client Name

Account ID

ACCOUNT

Account Type

Resolved many-to-many Relationship

PP

Role

ROLEClient IDAccount ID

Role Type

P P


Translating ER to Relational SchemaTranslating ER to Relational Schema

Each entity becomes a a tabletable (relation)

The attributes associated with each entity become attributesattributes of the relationThe key is nonnon--nullifiednullified attributesattributes

The keys of the associated entities with dependency are foreign keysforeign keys

SE352b 3-Mar-06



ER: ER: Identity & Functional DependencyIdentity & Functional Dependency

IDEF1X Notation


Account ID

ACCOUNT

Account Type

OwnClient ID (FK) Foreign Key

CLIENT

Nonprimeattributes

Client Name

Client ID


Translating to RelationsTranslating to Relations

CREATE TABLE CLIENT (Client_Number VARCHAR2(20) NOT NULL,Client_Name VARCHAR2(20) NULL,

);

CREATE TABLE CLIENT CLIENT (Client_Number VARCHAR2(20) NOT NULL,Client_Number VARCHAR2(20) NOT NULL,Client_Name VARCHAR2(20) NULL,

);

CREATE TABLE ACCOUNT (Account_Number VARCHAR2(20) NOT NULL,Client_Number VARCHAR2(20) NOT NULL,Account_Type VARCHAR2(20) NULL,

);

CREATE TABLE ACCOUNTACCOUNT (Account_Number VARCHAR2(20) NOT NULL,Account_Number VARCHAR2(20) NOT NULL,Client_Number VARCHAR2(20) NOT NULL,Client_Number VARCHAR2(20) NOT NULL,

Account_Type VARCHAR2(20) NULL,);


Account IDAccount ID

ACCOUNT

Account Type

OwnClient ID (FK)Client ID (FK) Foreign Key

CLIENT

Nonprimeattributes

Client Name

Client ID

SE352b 3-Mar-06



Database Management

System

Database SystemDatabase System

Database

Application

Data Design vs. ImplementationData Design vs. ImplementationER vs. Relational DatabaseER vs. Relational Database

Entity

Attribute

Instances

Relationship

Data Architecture(Logical Model)

Inv.Lee456

Reg.Smith123

Account Type

Client IDAccount ID

ImplementationImplementationDesignDesign

⇒ Table

⇒ Columns

⇒ Rows

⇒ Foreign Keys

Column

Row

ACCOUNT Table

Client ID

CLIENT

Client Name

Account IDACCOUNT

Account Type

POwnClient ID



Data Models

Schemas

Schema Architecture

DBMS Languages



Business Data Types



Query Processing


SE352b 3-Mar-06



Quality of DB DesignQuality of DB Design

Begin by discussing good and bad relation schemasschemas

Informal measures of the quality of relation schema design

Semantics of the attributes

Reducing the redundant values in tuples

Reducing the null values in tuples

Define normal forms formal measures of the quality of schemas

restrictions on the form of relation schemas


Manuf is attribute of Car and Not of OwnerCarManuf is attribute of Car and Not of OwnerCar

Semantics of the Relation AttributesSemantics of the Relation Attributes

The elements of the schema should be easy to explain

Keep attributes from different entities and relationships distinctExample

OwnerCar: (OId,Oname, DLNum, CarId, Make, Manuf)OIdOnameDLNumCarIdMakeManuf

OwnerCar

SE352b 3-Mar-06



Irrelevant to the Owner

Redundant InformationRedundant Information

Consider the following :(123,Joe, 123456789,106, Dodge, Chrysler)

(223,Moe, 223456789, 107, Dodge, Chrysler)

The Manuf attribute is redundant!

This leads to difficulty in updates (Update Anomalies)e.g. changing the Manuf for owner with OId 123 requires also changing it for OId 223.

OwnerCar: (OId, Oname, DLNum, CarId, Make, Manuf)OwnerCar: (OId, Oname, DLNum, CarId, Make, Manuf)


Update AnomaliesUpdate Anomalies

Insertion anomaliesWhen inserting a new owner,

we must correctly insert the Manuf, or will create inconsistencies

Cannot create a car without an owner

Deletion anomaliesDeletion of owner of a car

also deletes make and manufacturer of car

Deletion of owner of the last Dodge

deletes relationship between Dodge and Chrysler

Modification anomaliesChanging the make of a car requires consistency check

Cannot change so that a Dodge is made by Ford

OwnerCar: (OId,Oname, DLNum, CarId, Make, Manuf)OwnerCar: (OId,Oname, DLNum, CarId, Make, Manuf)

(123,Joe, 123456789,106, Dodge, Chrysler)

SE352b 3-Mar-06



X Y is a full functional dependencyif the removal of any attribute Ai from X removes the dependency

Not (X - {Ai} Y)

X Y is a partial dependencyif some attribute Ai may be removed without removing the dependency

X - {Ai} Y

X Y is a full functional dependencyif the removal of any attribute Ai from X removes the dependency

Not (X - {Ai} Y)

X Y is a partial dependencyif some attribute Ai may be removed without removing the dependency

X - {Ai} Y

Normalization:Normalization:Definition of Functional DependencyDefinition of Functional Dependency

Functional dependency

a constraint between 2 sets of attributes from the database

For each value of the first set there is a unique value of the second set

XX YY restricts the tuples that can be instances of the relation Rs.t. if t1 and t2 are instances of R

t1(X) = t2(X) then t1(Y) = t2(Y)For example, Client(CId, CName, CAddress)

{CId} {CName, CAddress}

A1A2

A3A4A5

R

X Y

} X

} YCId

CNameCAddress

Client


Normalization FormsNormalization Forms

A relation R (table) is in

1NF: if all attribute values are atomic (or indivisible)A relation that contains no repeating groups

2NF: If it is in 1NF and every nonprime attribute is fully functionally dependent on the primary key of R

3NF: if it is in 2NF andthere is no nonprime attribute that is functionally dependenton a non-key set of attributes

SE352b 3-Mar-06



UnUn--normalized Tablenormalized Table

Key List List List CompositeAccountant

NumberSkill

NumberSkill

Category Prof Accountant Name

Accountant Age

Office Number

Office City

Office Supervisor

21 113 Systems 3 Alice Adams 55 52 NYC Brown113 Systems 5179 Tax 1204 Audit 6

50 179 Tax 2 Charlie Cody 40 44 LA Green

148 Consulting 6179 Tax 6

LA Green

NYC Brown

35 Doug Baker

77 Doug Doe 52 52

32 44

Instances with Repeated GroupsInstances with Repeated Groups


1NF Table1NF Table

Accountant Number

Skill Number

Skill Category Prof

First Name

Last Name

Accountant Age

Group Number

Group City

Group Supervisor

21 113 Systems 3 Alice Adams 55 52 NYC Brown35 113 Systems 5 Doug Baker 32 44 LA Green35 179 Tax 1 Doug Baker 32 44 LA Green35 204 Audit 6 Doug Baker 32 44 LA Green50 179 Tax 2 Charlie Cody 40 44 LA Green77 148 Consulting 6 Doug Doe 52 52 NYC Brown77 179 Tax 6 Doug Doe 52 52 NYC Brown

Key





{Account Number, Skill Number} {Skill Category, Prof, …}{Account Number, Skill Number} {Skill Category, Prof, …}X Y

SE352b 3-Mar-06



2NF Table2NF Table

Key

Accountant Number

First Name Last Name

Accountant Age

Group Number

Group City

Group Supervisor

21 Alice Adams 55 52 NYC Brown35 Doug Baker 32 44 LA Green50 Charlie Cody 40 44 LA Green77 Doug Doe 52 52 NYC Brown

Accountant Table

KeySkill

NumberSkill

Category113 Systems179 Tax204 Audit148 Consulting

Skill Table

Accountant Number

Skill Number Prof

21 113 335 113 535 179 135 204 650 179 277 148 677 179 6

Proficiency TableKey




2NF Table2NF Table

Key

Accountant Number

First Name Last Name

Accountant Age

Group Number

Group City

Group Supervisor

21 Alice Adams 55 52 NYC Brown35 Doug Baker 32 44 LA Green50 Charlie Cody 40 44 LA Green77 Doug Doe 52 52 NYC Brown

Accountant Table



3NF: if it is in 2NF and there is no nonprime attribute that is functionally dependent on a non-key set of attributes


SE352b 3-Mar-06



3NF Table3NF Table

KeyGroup

NumberGroup

CityGroup

Supervisor44 LA Green52 NYC Brown

Group TableKey

Accountant Number

First Name

Last Name

Accountant Age

Group Number

21 Alice Adams 55 5235 Doug Baker 32 4450 Charlie Cody 40 4477 Doug Doe 52 52

Accountant Table

KeySkill

NumberSkill

Category113 Systems179 Tax204 Audit148 Consulting

Skill Table

Accountant Number

Skill Number Prof

21 113 335 113 535 179 135 204 650 179 277 148 677 179 6

Proficiency TableKey




NormalizationNormalization

Table with repeating groups

Table with repeating groups

1NF1NF

Remove repeating groupsRemove repeating groups

2NF2NF

Remove partial dependencyRemove partial dependency

3NF3NF

Remove transitive dependencyRemove transitive dependency

Each attribute relies on the key, the whole key, and nothing but the key

so help me codd!”

Each attribute relies on the Each attribute relies on the keykey, , the the whole keywhole key, and nothing but the , and nothing but the keykey

so help me so help me coddcodd!”!”

SE352b 3-Mar-06



Referential IntegrityReferential IntegrityIntegrity Constraints:

A set of (business) rules that address the validity of references by one object in DB to some other object(s) in the DB

Insertion Rule

A row should not be inserted in the referencing table unlessthere already exists a matching entry in the referenced table

Deletion Rule

A row should not be deleted from the referenced table if there is a matching row(s) in the referencing table

Enforcing Referential Integrity:Referential integrity options are declared when relational tables are defined

then, enforced by DBMS



Data Models

Schemas

Schema Architecture

DBMS Languages



Business Data Types



Query Processing


SE352b 3-Mar-06



SQLSQLStructured Query Language

ANSI and ISO standarde.g., SQL2 or SQL-92

SQL isa data definition language (DDL), and

a data manipulation language (DML), and

a programming (4th generation) language

We can use SQL forLogical database specification (database schema definitions)

Physical database specifications (indexes, etc.)

Querying database contents

Modifying database contents


SQL as SQL as DDLDDL::Create Table StatementCreate Table Statement

create table Customer (accountId int,lastName varchar(32),firstName varchar(32),street varchar(100),city varchar(32),state char(2),zipcode varchar(9)

);

Note that SQL has specific types

SE352b 3-Mar-06



Data types in SQLData types in SQL

integer integer, int, smallint, long floating point float, real, double precision

Numeric types

formatted decimal(i,j), dec(i,j)fixed length char(n), character(n) Character-string

types varying length varchar(n), char varying(n), character varying(n)

fixed length bit(n) Bit-string types varying length bit varying(n)

Date and time types

date, time, datetime, timestamp, time with time zone, interval

character long varchar(n), clob, text Large types binary blob


)


)


SQL as SQL as DMLDMLInsert Into <table> <first_column,...last_column> values<first_value,...last_value>;

Insert IntoInsert Into employee (first, last, age, address, city, state) valuesvalues('Luke', 'Duke', 45, '2130 Boars Nest', 'Hazard Co', 'Georgia');

Delete from <table> where <column> Operator <value> [and|or <column> Operator <value>]; [ ] = optional

Delete fromDelete from employee wherewhere lastname = 'May';

Update <table> set <column = newvalue>[<nextcolumn = newvalue2...] where <column> Operator <value>

[and|or <column> Operator <value>];[ ] = optional

UpdateUpdate phone_book setset area_code = 623 wherewhere prefix = 979;

SE352b 3-Mar-06



Relational Operations in SQLRelational Operations in SQLSelect statement

select <attribute names> from <tables>where <condition>

Projection in SQL using select clause

SelectSelect title fromfrom Movies

Selection in SQL using where clause

select * from Customer wherewhere lastName = 'Doe'

select distinctdistinct lastName, firstName from Customer

no duplicates with distinct


SQL as SQL as VDLVDLProducts and Joins in SQL vs. Products and Joins in SQL vs. ViewsViews

Cartesian product in SQL using from clauseSelect * from Employee, Timecard

Join using from and where clauses

Select * from Employee, Timecard whereEmployee.sin = Timecard.sin

Create View Test As

Create View EmployeeCard As

Select * from Employee, Timecard where Employee.sin = Timecard.sin;

Create View EmployeeCard As

Select * from Employee, Timecard where Employee.sin = Timecard.sin;

create table Employee (Sin int,lastName varchar(32),firstName varchar(32),street varchar(100),city varchar(32),state char(2),zipcode varchar(9)

)

create tabletable Employee (Sin int,lastName varchar(32),firstName varchar(32),street varchar(100),city varchar(32),state char(2),zipcode varchar(9)

)

SE352b 3-Mar-06



ThreeThree--Schema ArchitectureSchema Architecture(Cont.)(Cont.)

DATABASE

End User End User

External LevelExternal LevelExternal Schema

VIEWExternal Schema

VIEWExternal Schema

VIEWExternal Schema

VIEW

Conceptual LevelConceptual Level Conceptual SchemaConceptual Schema

Internal LevelInternal Level Internal SchemaInternal Schema

EmployeeCard

Employee


Views Views Example…Example…

Section Id

YearSemesterGradeCourse Number

Student TranscriptStudent NameTRANSCRIPT

Database

Data generated on the flyUsing subject areas entitiesData generated on the flyUsing subject areas entities

Brown

Smith

13599FallACS3380

10299SpringBCS3320

9298FallACS1310

8598FallAMATH2410

11299FallBMATH2410

11999FallCCS1310

Database Definition

(MetaMeta--DataData)

DepartmentCreditCourse NumberCourse NameCourse NameCOURSE

InstructorYearYearSemesterSemesterCourse NumberSection IdentifierSection IdentifierSECTION

MajorClassStudent NumberNameNameSTUDENT

SE352b 3-Mar-06



SQL: 4SQL: 4thth Generation PLGeneration PLNested QueriesNested Queries

Nested select query

Select videoId, dateAcquiredfrom Videotape where videoId in (

select videoId from Rental where dateRented=‘1/1/03’)

compare with

Select v.videoId, dateAcquiredfrom Videotape v, Rental rwhere v.videoId = r.videoId and

dateRented=‘1/1/03’)


Exists and Unique QueriesExists and Unique Queries

Find employees who have no time cards

Select firstName, lastName from Employee ewhere not exists

(select * from TimeCard t where e.sin=t.sin)

Find managers who have exactly one time card

select firstName, lastName from Employee ewhere unique

(select * from TimeCard t where t.sin=e.sin)and exists

(select * from Store where sin=manager)

SE352b 3-Mar-06



Aggregate FunctionsAggregate Functions

How many employees work at Store 3

select countcount (*) from WorksAt

where storeId=3

What is the average hourly rate of all hourly employees

select averageaverage(hourlyRate) from HourlyEmployees

How many different salaries are there?

select count (distinct salary) from SalariedEmployee



Data Models

Schemas

Schema Architecture

DBMS Languages



Business Data Types



Query Processing


SE352b 3-Mar-06



Database Management

System

Database SystemDatabase System

Data

ApplicationQuery ProcessingQuery Processing

How to represent relational DB?Each table is a file

File is a random access collection of records

Query is executed by reading records from files

Read record, create object in memory

Process object

Write result as a file of records or

keep in memory

Data file

4625

Byte address0

File contentsAnimal House&189&comedy&@Duck Soup&987&comedy&@Elizabeth&450&costume drama&@


IndexesIndexes

If selection queries are frequent, sorting the file or building indexes is important

Hash-based indexes only good for equality search

Sorted files and tree-based indexes best for range searchalso good for equality search

Data file

4625

Byte address0


Index to file

KeymovieId

Address

189450987

04625

SE352b 3-Mar-06



IndexesIndexes

An index on a file speeds up selections on the search key fields for the index

Any subset of the records of a relation can be the search key for an index on the relation

Search key is not the same as the relation key

An index contains a collection of data entries, and

supports efficient retrievalof all data entries with a given key value

Data file

4625

Byte address0


Index to file

KeymovieId

Address

189450987

04625


B+ TreeB+ Tree

B+ tree index: Balanced treesThe Most Widely Used Index

Ordered sequential file, linked list of ordered records

Good for range queriesSearch index for first record in range

Read data file sequentially

(linked list order) for rest of records

Insert/delete

at log f N cost;

(f = fanout, N = # leaf pages)

keep tree height-balanced

B+ treeP Y

A B C

· · · ·

Q S U W

· · · ·

X Y

· ·

H K M P

· · · ·

E G

· ·

C G P W Y

A ... B ... E ... G ...

Q ... S ...

C ... H ... P ...

W ...X ... Y ...Data file

First level

Third level

Second level

SE352b 3-Mar-06



Example: Processing a Range QueryExample: Processing a Range Query

select * from Customer where accountId >= 101 and accountId < 300select * from Customer where accountId >= 101 and accountId < 300

347 903

10 24 35

• • •

700 767 876 901

• • • •

902 903

• •

123 200 246 347

• • • •

101 110

• •

35 110 347 901 903

Data file

10 ... 24 ... 101 ... 110 ...35 ...

123 ... 200 ... 246 ... 347 ... 700 ... 767 ...

902 ... 903 ...876 ... 901 ...

B+ tree index for accountId


B+ Trees in PracticeB+ Trees in Practice

Typical order: 100. Typical fill-factor: 67%.average fanout = 133

Typical capacities:Height 4: 1334 = 312,900,700 records

Height 3: 1333 = 2,352,637 records

Can often hold top levels in buffer pool:Level 1 = 1 page = 8 Kbytes

Level 2 = 133 pages = 1 Mbyte

Level 3 = 17,689 pages = 133 MBytes

SE352b 3-Mar-06



TransactionsTransactions

A transaction is

a sequence of read and write operations on data items Logically, functions are treated as one unit of work

should either be done entirely or not at all

if it succeeds,the effects of write operations persist (commit);

if it fails,no effects of write operations persist (abort)

these guarantees are made despite concurrent activity in the system, and despite failures that may occur


Desirable Properties of Transactions: Desirable Properties of Transactions: ACIDACID

Atomicitythe property of a transaction that all of the updates are successful,or there is no update at all.

Consistencyeach transaction should leave the database in a consistent state

Properties such as referential integrity must be preserved

Isolationeach transaction when executed concurrently with other transactions should have the same affect as if it had been executed by itself

Durabilityonce a transaction has completed successfully, its changes to the database should be permanent.

Even serious failures should not affect the permanence of a transaction

SE352b 3-Mar-06



Transactions in SQLTransactions in SQL

A transaction begins when any SQL statement that queries the DB begins

To end a transaction,

the user issues

a Commit or

Rollback statement

OperationsBegin

Read

Write

Commit

Rollback


Example of TransactionExample of Transaction

Begin transaction

videoId video1 = select id of a copy of "Star Wars"

if (video1 == null) rollback transaction

insert row into Reservation for video1

videoId video2 = select id of a copy of "Return of the Jedi"



videoId video3 = select id of a copy of "The Empire Strikes Back"



commit transaction

Begin transaction

videoId video1 = select id of a copy of "Star Wars"



videoId video2 = select id of a copy of "Return of the Jedi"



videoId video3 = select id of a copy of "The Empire Strikes Back"



commit transaction

SE352b 3-Mar-06




Data Models

Schemas

Schema Architecture

DBMS Languages



Business Data Types



Query Processing



Interacting with Interacting with DBsDBs::Client Server ArchitectureClient Server Architecture

UserInterface Application

MiddlewareMiddlewareMiddleware

layer

DatabaseServer Database

Server

DatabaseServer layer

DatabaseServer

User and ApplicationClient layer

UserInterface Application

DBApplication

DBApplication

SE352b 3-Mar-06



Interacting with Interacting with DBsDBs::Java EnvironmentJava Environment

MiddlewareLayerMiddlewareMiddlewareLayerLayer

DatabaseServer LayerDatabaseDatabaseServer LayerServer Layer

Client LayerClient LayerClient Layer JavaJavaProgramProgram

java.sqljava.sqlpackagepackage

JDBC-ODBCbridge

ODBC databaseclient

Oracle Databaseclient

Middlewareclient

Middlewareserver

Oracleserver

Accessserver

Sybaseserver

Oracle Databaseclient

Sybase Databaseclient

Oracle ODBCclient

Access ODBCclient

JDBC packageJDBC package JDBC packageJDBC package JDBC packageJDBC package

Java Database Connectivity (JDBC)


Interacting with Interacting with DBsDBs::Java EnvironmentJava Environment

Package java.sql includes interfaces

Driver: supports the creation of a data connection

Connection: represents the connection between a Java client and an SQL database server

DatabaseMetaData: contains information about the database server

Statement: includes methods for executing text queries

PreparedStatement: represents a pre-compiled and stored query

CallableStatement: used to execute SQL stored procedures

ResultSet: contains the results of the execution of a select query

ResultSetMetaData: contains information about a ResultSet, including the attribute names and types.

SE352b 3-Mar-06



/* CIA.java/* CIA.javaFrom http://sqlzoo.net By Andrew CummingFrom http://sqlzoo.net By Andrew Cumming

*/*/import java.sql.*;import java.sql.*;public class CIA{public class CIA{public static void main(String[] args){public static void main(String[] args){Connection myCon;Connection myCon;Statement myStmt;Statement myStmt;try{try{Class.forName("com.mysql.jdbc.Driver").newInstance();Class.forName("com.mysql.jdbc.Driver").newInstance();# Connect to an instance of mysql with the follow details:# Connect to an instance of mysql with the follow details:# machine address: pc236nt.napier.ac.uk# machine address: pc236nt.napier.ac.uk# database : gisq# database : gisq# user name : scott# user name : scott# password : tiger# password : tigermyCon = DriverManager.getConnection(myCon = DriverManager.getConnection(

"jdbc:mysql://pc236nt.napier.ac.uk/gisq""jdbc:mysql://pc236nt.napier.ac.uk/gisq""scott","tiger");"scott","tiger");

myStmt = myCon.createStatement();myStmt = myCon.createStatement();ResultSet result = myStmt.executeQuery(ResultSet result = myStmt.executeQuery(

"SELECT * FROM cia WHERE population>200000000");"SELECT * FROM cia WHERE population>200000000");while (result.next()){while (result.next()){System.out.println(result.getString("name"));System.out.println(result.getString("name"));

}}myCon.close();myCon.close();

}}catch (Exception sqlEx){catch (Exception sqlEx){System.err.println(sqlEx);System.err.println(sqlEx);

}}}}

}}


#!C:/perl/bin/perl#!C:/perl/bin/perluse CGI qw(:standard);use CGI qw(:standard);use Mysql;use Mysql;

print header();print header();

#Import the CGI parameters into the $Q structure#Import the CGI parameters into the $Q structureimport_names();import_names();

#Connect to Andrew's MySQL server at Napier University#Connect to Andrew's MySQL server at Napier Universitymy $dbh = DBI->connect("dbi:mysql:database=gisq;host=pc236nt.napier.ac.uk;port=3306",my $dbh = DBI->connect("dbi:mysql:database=gisq;host=pc236nt.napier.ac.uk;port=3306",

"scott", "tiger");"scott", "tiger");

my $sth = $dbh->prepare("SELECT DISTINCT region FROM cia");my $sth = $dbh->prepare("SELECT DISTINCT region FROM cia");$sth->execute;$sth->execute;print "<h1>Here are some regions</h1>\n";print "<h1>Here are some regions</h1>\n";while (my ($region) = $sth->fetchrow_array()){while (my ($region) = $sth->fetchrow_array()){

print "<a href='?region=$region'>$region<a><br/>\n";print "<a href='?region=$region'>$region<a><br/>\n";}}

if ($Q::region) {if ($Q::region) {print "<h1>Here are the countries of $Q::region</h1>";print "<h1>Here are the countries of $Q::region</h1>";my $sth = $dbh->prepare("SELECT name, population FROM cia where region=?");my $sth = $dbh->prepare("SELECT name, population FROM cia where region=?");$sth->execute($Q::region);$sth->execute($Q::region);while (my ($name,$pop) = $sth->fetchrow_array()){while (my ($name,$pop) = $sth->fetchrow_array()){

print "$name $pop<br/>\n";print "$name $pop<br/>\n";}}

}}

$dbh->disconnect;$dbh->disconnect;

SE352b 3-Mar-06



Interacting with Interacting with DBsDBs::Web Server ArchitectureWeb Server Architecture

Web ServerMiddleware

layer

DatabaseServer Database

ServerDatabase

Server

User and ApplicationClient layer

WebBrowser

DBApplication

WebBrowser

WebBrowser

WebBrowser

DBApplication

FileSystems

DatabaseServer layer


Web Server Web Server ArchitecturesArchitectures

The architecture type depends on kind and number of servers involved

Different archictures different advantages and disadvantages

Generally we can distinguish between

2-tier architecture

3-tier architecture

n-tier architecture

What matters: SPEEDSPEED

SE352b 3-Mar-06



Advantages:easy and fast to setup

easy to administrate

Disadvantages:not fail safe (single point of failure)

scales badly on high loads

22--tier Architecturetier ArchitectureWeb server + module connecting to database

HTTP-Request

HTML-File Module

Directory

1

2

Web-Server

DB


33--tier Architecturetier ArchitectureWeb server + application server connecting to database

1

3

2

Web Server[Cluster]

Application Server[Cluster]

Other Servers[Cluster]

DBRepl.

DBDB DB

Advantages:better scalabilty

more reliable

through various mechanisms

offers better load balancing

Disadvantages:complicated to set up an maintain

Software Tools - Instructinstruct.uwo.ca/engin-sc/se352b/notes/HG-SE352Lect-W06... · 2006-03-04 · SE352b 3-Mar-06 (c) Hamada Ghenniwa, ECE, UWO 3 Feb 24, 2006 SE352b, ECE,UWO,

Documents