Transcript Database management

GI Systems and Science
January 30, 2012
Points to Cover
Recap of what we covered so far
 A concept of database

 Database Management System (DBMS)

Database data models
 Relational database model
 Object-oriented approach
Relationship between spatial and attribute data
in GIS
 Data management operations in ArcGIS

Spatial Data Modeling
Stage of data modeling in GIS
Available options
Indentifying the spatial features form
the real world that are of interest in
the context of the research question
Separating real world into layers of
feaures
Choosing how to represent the
features
Points
Lines
Polygons
Networks
Surfaces
Choosing an appropriate spatial data
model
Raster model
Vector model
Selecting an appropriate spatial data
structure to store the model within
the computer
Basic raster data structure
Compacted raster data structure
Basic vector data structure
Point dictionary structure
Topological vector data structure
Data Types in GIS
Spatial data
Attribute data
Concept of Database

A GIS can be described as a spatially-enabled
database
 Conventional databases store attribute data
 GIS stores both attribute and spatial data

Database is a set of structured data that are
related to each other in some way
 Organized filing cabinet
 Phone book
 On-line database of academic publications
Concept of Database

Many of human activities produce data that is
stored and managed in the database
environment
 Our society is generating tremendous amounts of
data on daily basis
 Data has become a valuable commodity
 To realize its value, the data need to be managed and
shared efficiently
Concept of Database

One of the approaches to turning data into
information is a database
 Database is more than just a container for storing
data
 Organize data into more meaningful and usable form
 Have analytical capabilities
Concept of Database

The database approach provides the following
benefits
 Ease of accessing data
 Prevention of unnecessary duplication of data
 Data stored independently of the application for
which they are used
 Secure controlled access to data
 Standards facilitate data exchange
 Data in the database can be shared by different users

Manual databases versus computer databases
 Which are more effective and efficient?
Database Management System

A DBMS is a computer
program that controls the
storage, retrieval and
modification of data in a
database (Dale and
McLaughlin, 1988)
 Comprises tools that
perform these functions
 Allows users to deal with the
data without knowing much
about the database itself
Figure 4.4
Source: Heywood et al., 2011
Relational Database Data Model

A DBMS manages data that are organized using
a database data model
 Set of rules about how the objects and relationships
between them should be represented
Number of different data models used for
handling attribute data in GIS
 Relational data model is most used by GIS

 The relational data model is based on concepts
proposed by Frank Codd (1970)
Relational Database Data Model
Data are organized and stored in a series of twodimensional tables, each of which contains records for
one type of entity
 Each entity has a unique identifier value assigned to it

 Unique identifiers allow to link (relate) data in two or more
different tables
 This structure makes possible applying queries to one or
more tables
Relational Database Data Model

Types of relationships possible between entities in a
relational database
Figure 4.7
Source: Heywood et al., 2011
Relational Database Data Model
Figure 4.8
Source: Heywood et al., 2011
Relational Database Data Model

Querying a relational database
 Queries are built on expressions based on relational
algebra which in turn is based on Boolean logic
 SQL, standard query language, has been developed to
facilitate the querying of relational databases
 Advantages: completeness, simplicity, pseudo-English
language style
 Disadvantages: was not developed to handle
geographical concepts such as ‘near to’, ‘far from’ or
connected to
Object-Oriented Database Approach

More realistic approach to representing spatial
entities in the database environment
 Encapsulation: Object = State + Behaviour
 State: set of values of attributes describing a spatial
entity
 Behaviour: methods of operating on it
 Composite object
 Hierarchy of objects
 Subclass
 Superclass
Figure 4.17
Source: Heywood et al., 2011
Object-Oriented Database Approach

Allows arranging objects into hierarchies allows
differential assignment of behaviours
(Inheritance)
 Behaviour of subclass objects = ‘own’ behaviour +
‘superclass’ behaviour

Benefits of the OO approach for GIS
 No differentiation between spatial and attribute data
 Works better for graphic operations

Disadvantage of the OO approach
 Represents world as a series of rigidly bounded
objects
 Still under development
Data management in ArcGIS

The primary data storage mechanism in ArcGIS
is the geodatabase
 A collection of geographic datasets of various types
held in a common ‘container’ such as a database file or
a database application
 Based on an object-relational model
 Relations (tables) function as objects
 Behavior is supplied through the geodatabase
application logic implemented as a series of system
tables

A key geodatabase strategy is to leverage the
database management system (DBMS)
 Extends SQL application to feature geometry
Data management in ArcGIS

Geodatabases comprise
 Three primary dataset types
 Feature classes
 Raster datasets
 Tables
 Database schema: metatables
Source: ArcGIS 10 Help files
containing information about object
behavior and relationships, maintaining
data integrity
 Topologies
 Networks
 Subtypes
*
Data management in ArcGIS

Types of geodatabases
 File geodatabases
 Stored as folders in a file system.
 Each dataset is held as a file that can scale up to 1 TB in
size.
 Personal geodatabases
 Datasets are stored within a Microsoft Access data file
 Limited in size to 2 GB
 ArcSDE geodatabases
 Stored in a relational database using Oracle, Microsoft
SQL Server, IBM DB2, IBM Informix, or PostgreSQL.
 Multiuser geodatabases which are unlimited in size
Data management in ArcGIS
Figure 4.9
Source: Heywood et al., 2011
Relationship between data in GIS

Raster datasets
 Simple raster datasets
 no separate attribute data table
 Rater datasets with attribute tables
 Within a geodatabase, the raster attribute table is saved
within the raster dataset and hidden from the user
Source: ArcGIS 10 Help files
*
Relationship between data in GIS

Vector datasets
Data Management Operations

The supported attribute column types in the
geodatabase
Source: ArcGIS 10 Help files
Data Management Operations

Queries
 Attribute query
 Used to find features based one particular attribute
 Locational query
 Used to find features with locations that meet certain
conditions
 Works with four types of relationships: near, adjacent to,
intersect, and inside
 The result of queries is a set of selected features
Data Management Operations

Joins
 Associating two or more tables based on a common
field (key)
One-to-one relationship
One-to-many relationship