fluency with information technology info100 and cse100 katherine deibel 2012-05-25katherine deibel,...

Designing DatabasesMore chaos… less chaos… red chaos… blue chaos

Fluency with Information Technology

INFO100 and CSE100

Katherine Deibel

2012-05-25 Katherine Deibel, Fluency in Information Technology 1

Monday is a Holiday

Ergo, no lab sections on Monday On Tuesday

Lab sections are optional

Attend any that you want

No attendance taken

Consider them to be "office hours"


Project 3B

Instructions now up Be sure to follow directions

Multiple files to turn-in

Two WebQ quizzes▪ First quiz can be taken multiple times

▪ Second quiz can be taken just once


Early Turn-In Extra Credit

Project 3B is due on Tueday, June 5 by 10pm… No late turn-ins!

Extra credit for early turn-in 4pts if by Monday, June 4 at 10pm

6pts if by Sunday, June 3 at 10pm

8pts if by Saturday, June 2 at 10pm

10pts if by Friday, June 1 at 10pm


Designing DatabasesMore chaos… less chaos… red chaos… blue chaos

Fluency with Information Technology

INFO100 and CSE100

Katherine Deibel


Making a Database

A database is made of … Physical database:

Tables actually stored on the hard disk

Logical databases:On-the-fly virtual tables created by queries

Queries:Commands written in SQL that define how to make a logical table from physical tables

How do we make each part?


Design Steps

Describe the data physically What are major clusters/higher-level

notions of the data?

How do they interact/relate with each other?

Identify the logical structure Describe a role with the data

Identify what fields are needed

Write queries for just those fields


Technologies for Design

There is a technology more powerful than computers when it comes to designing software, databases, etc.?

What is it? Pencil and paper

Whiteboards and markers

Blackboards and chalk


Step Away From The Machine

Why not start directly with the database software? Application locks you in to working in a

certain order

Paper-pencil gives you some flexibility

Application will force you to attend to the details and not the general goals


Database Design (on Paper)

Sketch relational structure using Entity-Relationship (ER) Diagrams Tables in boxes

Lines show relationship and type of relationship

Example: Specifying a 1-to-many relationship

FacultyStudent

∞1


Database Design (on Paper)

Outline key fields and refine the relationships with the fields


Faculty_IDFirstNameLastNameDepartmentDateHired

Faculty

Student_IDFirstNameLastNameMajorIDAdvisor

Student

∞

1

Database Design Paradigms

Build physical DB to avoid redundancy Each physical table represents one entity

from your ER diagram Expect that no physical table gives any

user their exact view To build view, build a query that

Joins tables together into a ‘super’ table

Trims out the items the user wants or needs


Defining Physical Databases Two aspects:

Underlying data architecture

Database schemes


Data Architecture

The underlying architecture shapes how the data is stored on the computer and how it is accessed Concerns efficiency in both speed and storage

space

Depends greatly on the purpose of and usage of the database

Not a concern for most users since DBMSs designers made these decisions already


Database Schemes

Database schemes (schema) are metadata specification that describes the database design

Includes Selecting data types (boolean, date, etc.)

Selecting formats (YYYY-MM-DD)


The Idea of Relationship

A relationship is a correspondence between rows of one table and the rows of another table

key Student_ID is used in each table, Find address for each student (Lives_At)

Find the student for each address (Home_Of)


Relationships in Practice


One-to-One Relationship

When one entry of one table can be linked with exactly one entry in another table

Example:Each US State has exactly one capitol city


StateIDNameCapitolCityJoinDatePopulation

US_State

CityIDNamePopulationMayorArea

City

1

1

One-to-Many Relationship

When one entry of one table can be linked with multiple entries in another table

Example: Faculty can advise multiple students


Faculty_IDFirstNameLastNameDepartmentDateHired

Faculty

Student_IDFirstNameLastNameMajorIDAdvisor

Student

∞

1

Many-to-Many Relationship When multiple entries of one table can be linked

with multiple entries in another table This is usually accomplished by using an

intermediary table Example:

A writer can be the author of multiple books and a book can have multiple authors


Book_IDTitleYearPublisher

Book

∞Writer_IDFirstNameLastName

Writer

1Authors_IDWriter_IDBook_ID

Authors1

∞

Designing a DatabaseThe Initial Steps


Katherine Deibel, Fluency in Information Technology 23

Premise

You have been contacted by a high school to build a database for the athletics program

What do you do?

2012-05-25

First Steps

CLIENTS

Staff Coach Players(?)

DATABASE DESIGNER

Interviews the experts/clients

Builds a list of design goals and rules for the database


Initial Statement of Goals

Individuals are selected for a team Keep track of the points awarded to

each student for participating for the awarding of school letters

The Database has to keep track of student Athletes over five years with any given Athlete participating in multiple sports in a given year.


Getting More Details

Each high school student can play for 1 to 4 years.

A player can play in 0, 1, or many sports each year

Each team has many players


More Specific Rules

The high school has many teams.

The high school is in one division.

Each team has a gender. Each team is in a division. Each team has a coach. Each team has many

players. Each student earns points

for winning an event. Each student earns a

letter by earning x points in a sport in a season.

Each student may participate in zero, one, or many sports each year.

Each student may play in sports for 1-4 years.

Each student may play in zero, one, or many competitions at an event.

Each coach can coach one or many teams.


Physical Database

Major data entities Teams

Athletes


Fields

The HS has many teams The HS is in one division Each team has a gender Each team is in a

division Each team has a coach A team has many

players Each student earns

points for winning an event

team_name n/a team_gender n/a coach_name one-to-many student_points

team_winsevent_nameevent_outcome


Physical Database

Major data entities Teams

Athletes

Events


More Fields

student_id

student_result_at_event

student_points_to_date

student_points_at_event

student_first_name

letters_sport_code

student_middle_name

letters_awarded_date

student_last_name

team_gender

student_date_of_birth

event_location

team_name

student_gender

team_description

student_address

coach_name

student_other_details

team_other_details

division_description

sport_description

event_name

event_start_date

event_end_date

event_other_details 2012-05-25 Katherine Deibel, Fluency in Information Technology 31

Process Continues

You keep adding More fields

More relationships Perhaps you discover the need more

entities (tables) Once complete, you implement the

tables And then the logical begins


Summary

Designing a database is more about thinking than typing Sketch out the relationships on paper

Identify the needs of the clients Key aspect is to separate the logical

and physical aspects Physical is about entities and their

relationships

Logical is about the user


Monday is a Holiday

Ergo, no lab sections on Monday On Tuesday

Lab sections are optional

Attend any that you want

No attendance taken

Consider them to be "office hours"


fluency with information technology info100 and cse100 katherine deibel 2012-05-25katherine deibel,...

Documents

information technologyinfo100

information technology2project

information technology1monday

information technology5making

information technology8step

physical database

redundancyeach physical

physical tableshow