Business Database Management Systems

1
Business Database Management Systems

• Class 2
• DSC 544/444
• Fall 2006

2
Agenda

• What business information should go into a
  database?
• Databases vs. spreadsheets
• A brief history of DB-time

3
What business information should go into a
database?

• Lets consider a specific company
• Alpine Ascents International,
• based in Seattle, and one of the worlds premier
  mountain-climbing guide services.
• As owners/managers of this business, lets
  brainstorm what kinds of information we will be
  juggling in this firm.

4
What business information should go into a
database?

• Given all the information we are juggling, what
  should be put into the database? All of it?
• What kind of information is appropriate for a DB?
• After we have decided what the database should
  store, there is the question of how we want it
  stored.
• What are our goals why do we care how it is
  stored?

5
Agenda

• What business information should go into a
  database?
• Databases vs. spreadsheets
• A brief history of DB-time

6
Databases vs. spreadsheets

• Most databases (not all, as we will soon see) are
  simply a collection of tables
• What does a database table look like?

7
Databases vs. spreadsheets

• Of course, most databases have many tables

8
Databases vs. spreadsheets

• Likewise, a single spreadsheet workbook file
  can hold multiple tables

9
Databases vs. spreadsheets

• A database is a collection of 2-D tables
• A spreadsheet workbook is a collection of 2-D
  tables
• What the difference between the two?
• From a design perspective, nothing!
• You can design database tables in Excel
• Most database designs start that way
• You can equally well design on paper

10
Databases vs. spreadsheets

• Once you have your tables created, there is a big
  difference in what the system can do for you
• Excel constructs (such as relative and absolute
  () references) for connecting information within
  a workbook are painful
• We will see an example of just how painful it is
  soon!
• Popular database systems support powerful
  constructs for connecting information across
  tables
• Relational databases Relational algebra
• relatively few concepts, but somewhat deep
• We will learn it!

11
Databases vs. spreadsheets

• Besides making it very painful to relate data
  between tables, when your table designs are put
  into use, spreadsheets (and paper, for that
  matter) arent good at addressign
• Security
• Maintainability
• Multiple Users
• Networkability
• Controlled Data Entry

12
Agenda

• What business information should go into a
  database?
• Databases vs. spreadsheets
• A brief history of DB-time

13
A brief history of DB-times

• Emergence (1890 1949)
• Manual processing used in tracking library
  records and tax payments
• First punched-card machine processed the censuses
  of 1890 and of 1900
• IBM developed special collating equipment to keep
  records on 26M individual employees
• Initially established with government support,
  the punched-card data processing industry became
  a major factor in U.S. companies later dominance
  in electronic computing

14
A brief history of DB-times

• First Generation (1950 - 1960)
• Data on paper tapes, punched cards or magnetic
  tapes
• Offline services / batch processing
• Sequential access to data

15
A brief history of DB-times

• Second Generation (1960 - 1965)
• First general-purpose database management system,
• GE (General Electric) develops Integrated Data
  Store system (IDS)
• Term database emerged
• 1962 launch of Sabre airline reservation system,
  from IBM and American Airlines

16
A brief history of DB-times

• Third Generation (1965 - 1975)
• Two competing data models
• Hierarchical
• Information organized using tree structures
• IMS (Information Management System) by IBM
• Network
• More complex relationship paths than trees
• First widely-accepted model, many mainframe-based
  implementations
• Being gradually phased out.
• 1970 E.F. Codd proposed the relational model for
  database in his landmark paper, A Relational
  Model of Data for Large Shared Data Banks.

17
A brief history of DB-times

• Fourth Generation (1975 - today)
• late 1970s
• SQL (Structured Query Language) definedhas now
  become the industry (and ISO) standard.
• Peter Chen proposed the Entity-Relationship (ER)
  model for database design giving another
  important insight into conceptual data models.
• Major relational systems introduced DB2 by IBM,
  Oracle by Oracle

18
A brief history of DB-times

• 1980s-1990s
• Client-server and parallel processing
  applications developed
• Web/DB use grows exponentially
• Increasingly complex products focus on client
  tools for application development
• e.g., Oracle Developer
• Object-oriented (and Object-Relational hybrid)
  databases develop

19
A brief history of DB-times

• Current Trends
• Faster hardware and more sophisticated software
• permits small companies to cost-effectively
  design and implement data-driven websites
• Web//DB integration tools
• E.g., Microsoft Visual Studio, Macromedia Dream
  Weaver, etc.
• Huge (terabyte) systems by major corporations
• Increased data-mining and data-warehousing
  applications

20
Where we commonly see databases used today

• Libraries
• Computerized checkout system in supermarkets
• Travel reservation systems
• Banks and stock exchanges
• Supply-chain management
• WWW search engines