Issues, Trends and Strategies for Computer Systems Management

1
Issues, Trends and Strategies for Computer
Systems Management
Chapter 6. Database Systems

• UMUC Graduate School of
• Management and Technology

2
Agenda

• Background
• Database Models
• Database Architecture
• Exercise

3
Background
4
Definitions

• Data - a representation of facts, concepts or
  instructions in a formalized manner suitable for
  communication, interpretation and processing
• Database - a collection of data gathered for a
  particular purpose or purposes
• Database management system (DBMS) - a software
  system that is used to define, create, store,
  maintain and manipulate the database
• Database system - the database, the DBMS, and
  application programs

5
Brief History

• 1940s - 50s Use of computers as calculators
• 1960s Business uses file system-based
• 1970s Relational model
• 1980s Microcomputer revolution local area
  networks
• 1990s Internet and the World Wide Web

6
Database Applications

• Transactional - airline reservation system,
  customer tracking system, inventory system,
  on-line transactional processing (OLTP)
• Decision Support - data warehousing, data marts,
  executive information systems, on-line analytical
  processing (OLAP)

7
Purpose of Database Systems
CRUD

• Collect
• Organize
• Store
• Retrieve
• Manipulate
• Output
• Recovery

8
Components of a DBMS
Design Tools
Run Time
DBMS Engine
Physical Management
Metadata
Database
9
Advantages of DBMSs

• Controls redundancies, inconsistencies
• Provides data sharing
• Supports data integrity
• Supports standardization
• Allows for data recovery
• Enforces data security
• Improves maintenance

10
Disadvantages of DBMSs

• Expensive, high overhead cost
• Complex
• May not be suitable for some applications
• Single point of failure possible security
  breaches

11
Database Models
12
Database Models

• Flat Files
• Hierarchical Databases
• Network Databases
• Relational Databases
• Object-Oriented Databases
• Object-Relational Databases

13
Flat Files (File-based)

• Collection of individual files accessed by
  applications programs
• Limitations
• Separate and isolated data
• Data duplication, inconsistencies
• Application program dependencies
• Incompatible files
• Lack of data sharing
• Training

Registrar
Class programs
Class file
Accounting
Accounting programs
Accounts file
Facilities
Facilities programs
Facilities file
14
Hierarchical Databases

• Relates data by rigidly structuring data into an
  inverted tree

Sales
Parent
West Coast
East Coast
Child
Sneakers
Shirts
Shorts
Sneakers
Shirts
Shorts
15
Network Databases

• Supports relationships among the data through
  linked list structure in which subordinated
  records (members) can be linked to more than one
  parent (owner).

Sales
Owner
Owner
West Coast
East Coast
Sneakers
Shirts
Shorts
Members
16
Relational Databases

• Emerged during late1970s and1980s
• Collection of tables, each with a primary key
• Relationships between tables established through
  a common attribute

17
Relational Databases
Field
Table
Employee
EID Name Title Age Division 173 Smith
, A. Dir., Acctg. 45 1001 275 Jones, S. Dir.,
TQM 32 1002
Record
Office
Division Address City 1001 100 Main
Street Boston 1002 200 Elm Street NY
Use Structured Query Language (SQL) to access and
manipulate data
18
SQL General Form

• SELECT columns
• FROM tables
• JOIN link columns
• WHERE conditions
• GROUP BY column
• ORDER BY column (ASC DESC)

19
SQL Introduction
Customer
C Name Phone City AccountBalance 28764 Adamz 602-
999-2539 Phoenix 197.54 87535 James 305-777-2235 M
iami 255.93 44453 Kolke 303-888-8876 Denver 863.39
29587 Smitz 206-676-7763 Seattle 353.76
Query Which customers have balances greater
than 200?
SQL SELECT C, Name, Phone, City,
AccountBalance FROM Customers WHERE AccountBalan
ce 200
Query In alphabetical order, which Denver
customers have balances greater than 200? Show
customer no., city and balance.
SQL SELECT C, City, AccountBalance FROM Custome
rs WHERE AccountBalance 200 and City
Denver ORDER BY Name ASC
20
QBE Example
21
QBE Example
Query Which customers have balances greater
than 200?
Query In alphabetical order, which Denver
customers have balances greater than 200? Show
customer no., city and balance.
22
Relational Databases

• Advantages
• Conceptual simplicity
• Flexibility in creating links among data
• Data independence -- logical design of the data
  is independent of the physical implementation
• Popular, fairly easy to use
• Disadvantages
• Design of large scale databases may be complex
  resulting in slower search and access times
• Data redundancies possible (correct by
  normalization)

23
Objects

• Hypertext Massive text
• Pictures Graphs
• Objects
• Video
• Sound
• User defined

24
Sample OO Database
Patient Visits ID Date Physician Problems Commen
ts
Patient Treatments ID Date Procedure Doctor
25
Object-Oriented Database Systems

• Originated from Object-Oriented Programming
  Languages
• SIMULA 67-Smalltalk-80
• Object-Oriented Databases Prototypes (late 80s
  early 90s)
• ORION (Microelectronic and Computer Technology
  Corporation), OpenOODB (Texas Instruments), IRIS
  (Hewlett-Packard), ODE (ATT Bell Labs),
  ENCORE/Observer (Brown University)
• Object-Oriented Database System(early to late
  90s)
• GEMSTONE/OPAL (ServioLogic), ONTOS (Ontologic),
  Objectivity (Objectivity Inc), Versant (Versant
  Technologies), ObjectStore (Object Design), O2
  (O2 Technology)

26
Object-Oriented Databases

• OODB Applications
• Computer-aided design and manufacturing (CAD/CAM)
• Computer-integrated manufacturing (CIM)
• Computer-aided software engineering (CASE)
• Geographic information systems (GIS)
• Science and medicine
• Document storage and retrieval

27
Object-Oriented Databases

• Objects, Methods, and Messages
• Everything is an Object
• Every Object has a type -- its Class
• Individual Objects are Instances
• Encapsulation
• internal structure of a given object not visible
  to the users of the object

28
Object-Oriented Databases

• Objects have unique identity
• Identity represented by Object ID (OID)
• Objects have a hierarchy
• Objects inherit from their parents
• Integrates a DBMS with the capabilities of an OO
  programming language
• Handles complex structures as objects (CAD/CAM
  applications)
• No need to read data and translate to application
  data structures

29
Object-Oriented Databases

• Advantages
• Handles abstract types well
• Better modeling of the real world
• Reusability
• Portability
• Disadvantages
• Performance
• Immature
• Small market
• Complex
• No theoretical foundation
• Less relevant for business applications

30
Object-Oriented Databases

• Too soon to know if object-oriented databases
  will become dominate currently maintains a small
  market
• They will have their niche
• CAD/CAM
• Relational database vendors fighting back
• Migrating to Object-Relational Database Systems

31
Object-Relational Databases

• Supports Abstract Data Types (multimedia objects)
  with search algorithms
• A field can contain a collection of multiple
  values or composite data types
• Likely to replace pure RDBMS

32
Object-Relational Databases

• Advantages
• Way to migrate to object-oriented systems
• Current vendors heading in this direction
• Improved performance
• Disadvantages
• More an extended relational model than
  object-oriented
• Complex
• Higher storage requirements than RDBMS

33
Future of Database Systems

• Object-relational approach will replace purely
  relational products
• Access to nontraditional data types
• Lite version of the DBMS for PDAs
• Advanced indexing will continue to improve query
  performance

34
Database Architecture
35
Key Aspects of a Database Architecture

• Data Model - the way the data is conceptually
  structured (relational, object-oriented,
  hierarchical, etc.)
• Data Definition Language (DDL) - defines what
  types of information will be recorded in the
  database
• Data Manipulation Language (DML) - language used
  to query the database and to store or update
  information in the database

36
Levels of Data Abstraction

• Many views, single conceptual schema, single
  physical schema

View
View
View
Subschema is part of the database as seen by a
user or application
Conceptual Schema
Internal/Physical Schema
Schema is a description of a particular
collection of data using a given data model the
structure of database including data type,
relationships and constraints
Database
37
Data Independence

• Insulates the application from the method used in
  storing and structuring the data
• Logical data independence - protects from changes
  in the logical structure of data
• Physical data independence - protects from
  changes in the physical structure of data

38
Database Development Process

• User needs assessment and requirements gathering
• Data modeling
• Implementation
• Testing
• Deployment
• Maintenance

39
Key Players in Database Management

• Users
• Data Administrator
• Database Administrator
• Database Designers
• Application Programmers

40
Student Exercise

• Selecting the Optimal Data Model"

41
Exercise

• Susan is employed in a small but progressive
  division and she was asked to quickly develop a
  personnel security database. This database is to
  contain such information as name, address,
  reference data, background checks, fingerprints
  and digital photographs.
• Fred maintains a high speed massive transactional
  system which tracks the value of government
  securities. Each minute, thousands of clients
  query this database via the web.
• Ann-Marie is the project chief of an
  enterprise-wide data warehouse project in her
  organization. Data from four legacy systems will
  be combined into a single database maintained by
  Ann-Marie's division.

42
Exercise

• Tony has a Rolodex-type card system on his desk.
  He wants to automate this system on his Personal
  Digital Assistant.
• Carrie is establishing a document management
  tracking system. This system would capture the
  title, creation date, revision dates, author,
  keywords and graphics associated with each
  document. Only specific queries can be made
  against each document.
• Sam works in a highly stovepiped organizational
  structure which is responsible for reporting
  field performance data to the regional offices,
  which in turn consolidates data from other field
  offices and sends it directly up to the
  Headquarters Office.

43
Exercise

• Marlene works in an organization with many
  cross-functional teams. She has been asked to
  track these teams in terms of name,
  responsibilities, team leader, participants,
  organizations represented by the participants,
  phone numbers, e-mail addresses, and organization
  codes.
• Chris is in charge of developing a Computer-Aided
  Software Engineering (CASE) database, which
  stores data relating to the stages of the
  software development lifecycle.
• Betty works in a multimedia publishing
  organization for the DoD and wants to track text,
  audio, image, and video data and animation.
• Mark wants to track secret information pertaining
  to military satellites, monitored countries, and
  tracking stations.