Data Management Options

1
Data Management Options

• Dr. Merle P. Martin
• MIS Department
• CSU Sacramento

2
Acknowledgments

• Dr. Russell Ching (MIS Dept) Source
  Materiel / Graphics
• Edie Schmidt (UMS) - Graphic Design
• Prentice Hall Publishing (Permissions)
• Martin, Analysis and Design of Business
  Information Systems, 1995

3
Agenda

• Why manage data?
• Definitions
• Typical problems
• Data Administrator
• The DBMS
• Distributing data

4
Why Manage Data?

• Delayed output
  (paycheck)
• Locate a resource
• Where is the stock
  item stored?
• Where does the employee
  work?

5
Why Manage Data?

• Make resource decisions
• Should we turn account over to
  collection agency?
• Should we send customer letter
  asking why he / she hasnt shopped
  here in 6 months?
• Should we give employee overtime?

6
Why Manage Data?

• Determine resource status
• Is there enough stock in warehouse to
  satisfy this customers order?
• How much should I order?
• What is the value of
  this resource?
• balance sheet

7
Definitions

• File resource inventory
• Material
• People
• Employees, customers
• Funds
• Customer balances
• Accounts Payable

8
Definitions

• Data Organization
• Bit / byte
• Character
• Field
• Record
• File
• DBMS

9
Data Hierarchy for Stereos to Go
Database

12345
Smith
John
A
123 Main Street
Sacramento
CA
95819
File
12345
Smith
John
A
123 Main Street
Sacramento
CA
95819
12345
Smith
John
A
123 Main Street
Sacramento
CA
95819
Record
Field
Smith
Character
10110011
(Byte)
Bit
1
10
Definitions

• Views
• Physical - how stored
• Logical - how viewed and
  used
• Volatility - records that change
• Immediacy rapidity of change

11
Storage Problems

• Redundancy
• Accuracy
• Security
• Lack of data sharing
• Report inflexibility
• Inconsistent data definitions
• Too much data
• information overload

12
Data Administrator

• Clean up data definitions
• Control shared data
• Manage distributed data
• Maintain data quality

13
Clean Up Definitions

• Synonyms / aliases
• Standard data definitions
• names and formats
• Date of Birth (AJIS)
• mm/dd/yy (courts)
• dd/mm/yy (corrections)
• Data Dictionary
• COBOL

14
Control Shared Data

• Local - used by one unit
• Shared - used by two
  or more activities
• Impact of proposed program changes on
  shared data
• Program-to-data element matrix
• Control or clearinghouse?

15
Manage Distributed Data

• Geographically dispersed
• whether shared data
  or not
• Different levels of detail
• different management levels

16
Aggregate
Infrequent
Quite old
External
Future
Wide
Low
StrategicPlanning
Management
Currency
Frequency of Use
Source
Scope
Time Horizon
Required Accuracy
Level of Aggregation
Control
Operational
Control
High
Internal
Detailed
Historical
Well defined
Very frequent
Highly current
17
Maintain Data Quality

• Put owners in charge
  of data
• verify data accuracy and
  quality
• Fairbanks Court example
• Who owns the data?

18
Issue

• Should the Data
  Administrator control
  ALL data,
• or just that data that crosses organizational
  boundaries?
• WHAT DO YOU THINK?

19
The DBMS

• Data Base Management System
  software that permits a firm
  to
• centralize data
• manage them efficiently
• provide access to applications
• such as payroll, inventory

20
DBMS Components

• Data Design
  Language (DDL)
• Data Manipulation Language
  (DML)
• Inquiry Language (IQL)
• Teleprocessing Interface (TP)
• Martin, Figure 16-5

21

Designers
Teleprocess
DDL
Database
IQL
DML
Interface
Update Retrieve

Applic. Software
End-Users
Programmers
22
IQL LANGUAGE
IQL SELECT EMP-ID, EMP-FIRST-NAME,
EMP-LAST-NAME, EMP-YTD-PAY FROM
EMPLOYEE WHERE EMP-ID1234 .

Data
Base
23
3-level Database Model

• James Martin
• Sprague / McNurlin,
  Fig. 7-2, pg. 207

24
External Level (1)

• User views (logical)
• By application program
• Each has unique view
• Schema / subschema

25
Schema and Subschemas
Physical Database
DBMS Software
DBMS
Overall View of the Data
Schema
Individual Views
Subschema
Subschema
Subschema
User
User
User
User
User
User
26
Enterprise Level (2)

• Under control of Data Administrator
• DBMS
• Implementation data removed
• passwords
• report views

27
Physical Level (3)

• Schema
• Pointers
  (e.g., next record)
• Flags
  (e.g., record frozen)

28
Traditional Data Models

• Hierarchical - one parent
• Network
• more than one parent
• student to course, major
• Relational (tables)

29
Hierarchical Model
Project 1
Dept. A
Dept. B
Dept C
1
3
5
2
4
6
Employees
30
Network Model
John Smith
Jane Smith
Savings
Mortgage
Checking
31
Relational
Account Number
Last Name
First Name
Credit Limit
Middle Initial
. . .
Customer
Order Number
Order Date
Account Number
Date Shipped
Orders
Order Number
Line Item Number
Product Code
Quantity
Line Items
Product Code
Product Name
Unit
Price
Manufacturer Code
Products
Manufacturer Name
Manufacturer Code
Manufac(turer)
32
Object-oriented DBMS

• An object is
• a piece of data PLUS
• procedures performed on
  data PLUS
• attributes describing data PLUS
• relationship between object and
  other objects

33
Distributed Data

• Goals
• move processing as close to
  users as possible
• allow several applications to run simultaneously
  on same data

34
Distributed Types

• Fragmented
• distribute data without duplication
• users unaware of where
  data located
• Segmented
• data duplicated
• one site has master file
• problem with data synchronization

35
Why Distribute?

• Save money
• offload DB processes to
  less expensive machines
  (PCs)
• Lower telecommunications costs
• DB closer to users
• Decrease dependence on a single computer
  manufacturer

36
Why Distribute

• Move control closer
  to owner
• Increased DBMS scope
• more varied types of data
• link at workstations
• Permit storage of multimedia data

37
True Distributed DB

• Local autonomy
  (ownership)
• No reliance on central site
• Continuous operations
• not affected by another site
• Data transparency
• Independence

38
Independence

• Fragmentation
• Replication
• Hardware
• Software
• Networks
• Database

39
Problems With Distributed Databases

• Security
• Shared data
• simultaneous update
• Complexity
• Need telecommunications infrastructure

40
Issue

• Is data in your organization totally
  distributed?
• How?
• Should it be?
• Why or why not?

41
Points to remember

• Definition
• Typical problems
• Role of Data Administrator
• The DBMS
• Distributing data