Benjamin PostCole Kelleher

Databases and GIS

Database Characteristics

1. Availability

Data must maintain a specified level of availability to the users

2. Performance

Database requests must be executed within time thresholds

3. Isolation between users

Concurrency

Commits/Rollbacks

4. Data Independence

Reduces redundancy, storage requirements

5. Self-describing

Able to describe the stored data using schemas

Comprised of relations between individual pieces of data

Concepts:

Table

Also known as relation, contains attributes and tuples

Attributes

Columns, stores the individual pieces of data

Tuples

Rows, forms the relationship between the data in the columns.

MUST BE UNIQUE ENTRY

Relational Databases

Diagrams Entities and their Attributes, and the Relationship

between entities.

Entity-Relationship Model

Employee Office BranchWorks at

NameStart DateSalary

AddressOffice HoursOffice Phone

1 N

Comprised of objects stored in memory

Concepts:

State

Defines information about the object, i.e. geographic location

Behaviors

Actions that the object can perform or actions which modify an object

Inheritance

Similar to Object-Oriented Programming, objectscan inherit state and behavior

from other objects

Object-Oriented Databases

Built on set theory and relational algebra

Data definition language

SQL can be used to define and create the database

Data manipulation language

INSERT

UPDATE

DELETE

Queries via SELECT

Structured Query Language (SQL)

Geometries in Oracle Spatial are split into two data types:

1. SDO_GEOMETRY• Stores the geometry as is.• Requires a larger storage space due to redundancy.• Useful for queries that retrieve the entire geometry of a feature.

2. SDO_TOPO_GEOMETRY• Stores geometry using topological elements.• Allows for topology management over multiple features.• Decreased redundancy and thus decreased storage space.• Supports topological queries and processing

Oracle Spatial is an open format and is compliant on a large number of platforms.Geometry data types are restricted to two dimensions.

Oracle Spatial – Geometries and Rasters

GeoRaster in Oracle Spatial is used to handle raster data.

• Uses its own “GeoRaster” data type to standardize the rasters stored in the database instead of storing the variety of raster file types that exist.

• Employs the SDO_GEORASTER data type for the storage of the raster data.• Also uses the SDO_GEOMETRY data type to store the spatial extent of the

raster data for ease of indexing. • This is a logically layered multidimensional data model that stores each

raster band as a layer.• Further development needed in geoprocessing and modeling capabilities.

Post GIS is an extension of PostgreSQL for spatial databases.• Certified under the OGC as a “Simple Features for SQL” database.• Supports spatial joins in which two tables are joined on a spatial relationship.• OGC compliancy means it is applicable to a large quantity of open source

software and some proprietary software.• Further development is needed for three dimensional modeling and raster

integration.

PostGIS

GRASS is an open source GIS package that can conduct spatial analysis and modeling.

• Supports both vector and raster data with three dimensional capabilities.

• Is able to interact with a variety of useful software packages to aid statistical operations, spatial analysis and processing.

• GRASS is OGC compliant which means its source code is available for the user to manipulate to fit his or her needs and contribute to the software itself.

• High level interoperability means an increase in flexibility and performance.

• GRASS is able to connect to a variety of RDBMS for ease of integration.

Geographic Resources Analysis Support Software (GRASS)