Spatio-Temporal Database

Doctoral Course: Conceptual Modeling

Januray 25, 2005
Jiyong Zhang([email protected])
Ecole Polytechnique Fdrale de Lausanne (EPFL)
School of Computer and Communication Sciences
CH-1015 Lausanne, Switzerland
1

Recent News: U.S. Submarine hit undersea mountain

Time: January 8, 2005

The accident occurred about 350 miles south of Guam
1 sailor died, 24 injured

The undersea map was

drawn on 1989
San Francisco, the nuclearpowered attack submarine
2

Jiyong Zhang

Januray 25, 2005

Outline

Introduction
Spatial Database
Temporal Database
Spatio-Temporal Database
Open issues for STDB Systems
Summary

Jiyong Zhang

Januray 25, 2005

Introduction

Spatio-Temporal Database:

A database that embodies spatial, temporal, and spatiotemporal

database concepts, and captures spatial and temporal aspects of
data
Dealing with geometry changing over time

Jiyong Zhang

Januray 25, 2005

Introduction (contd)

Example of Spatiotemporal Applications

Year 1980: landparcel A has common borders with B; river R

runs through A and B; A has soil type clay, B has soil type
forest
Year 1990: A was divided into A and A, B has soil type highdensity forest, A has soil type sparse forest
Year 1995: river R changed its position and become R

Jiyong Zhang

Januray 25, 2005

Spatial Database: Introduction

Many applications in various fields require management of

geometric, geographic or spatial data (data related to space)

The Common challenge:

A geographic space: surface of the earth

Man-made space: layout of VLSI design
Model of the human brain
3-D space representation of the chains of protein molecules

Dealing with large collections of relatively simple geometric

objects: e.g., 100,000 polygons

Jiyong Zhang

Januray 25, 2005

Spatial Database: Definition

A spatial database system:

Is a database system (with additional capabilities for handling

spatial data)
Offers spatial data types (SDTs) in its data model and query
language

Supports SDT in its implementation

Structure in space: e.g., POINT, LINE, REGION

Relationships among them: e.g., a intersects b
Spatial indexing: retrieving objects in particular area without scanning
the whole space
Efficient algorithm for spatial joins

Jiyong Zhang

Januray 25, 2005

Spatial Database: Modeling

Assume 2-D GIS application, two basic things need to be

represented:

Objects in space: cities, forests, or rivers

distinct entities arranged in space, each of which has its own
geometric description
=>modeling single objects
Space: describe the space itself
say something about every point in space
=>modeling spatially related collections of objects

Jiyong Zhang

Januray 25, 2005

Spatial Database: Modeling (contd)

Fundamental abstractions for modeling single objects:

Point, Line, Region

Spatially related collections of objects:

Partition, Networks

Jiyong Zhang

Januray 25, 2005

Spatial Database: Spatial Data Types and Operations

A sample System: ROSE (Guting and Schneider, 1993)

Three data types:

Define two type sets:

Points, lines, regions

EXT={lines, regions}, GEO={points, lines, regions}

Four classes of operations:

1. Spatial Predicates for topological relationships:

10

Jiyong Zhang

Januray 25, 2005

Spatial Database: Spatial Data Types and Operations (contd)

2. Operations returning atomic spatial data type values:

3. Spatial operations returning number:

4. Spatial operations on set of objects:

11

Jiyong Zhang

Januray 25, 2005

Spatial Database: Spatial relationships

Topological relationships

Direction relationships

Above, below, north_of, southwest_of,

Metric relationships

12

Disjoint, touch, overlap, in, cover, equal

Distance

Jiyong Zhang

Januray 25, 2005

Spatial Database: Querying

Two main issues:

1. Connecting the operations of a spatial algebra to the facilities

of a DBMS query language.
2. Providing graphical presentation of spatial data (i.e. results of
queries), and graphical input of SDT values used in queries.

13

Jiyong Zhang

Januray 25, 2005

Spatial Database: Querying (contd)

Fundamental spatial algebra operations:

Spatial selection: returning those objects satisfying a spatial

predicate with the query object

Spatial join: A join which compares any two joined objects

based on a predicate on their spatial attribute values

14

Example: All big cities no more than 300Kms from Lausanne

SELECT cname FROM cities c WHERE dist(c.center,
Lausanne.center) < 300 and c.pop > 500K

For each river pass through Switzerland, find all cities within less
than 50KMs
SELECT c.cname FROM rivers r, cities c
WHERE r.route intersects Switzerland.area and
dist(r.route, c.area) < 50KM

Jiyong Zhang

Januray 25, 2005

Spatial Database: Querying (contd)

Requirements for spatial querying

15

Spatial data types

Graphical display of query results
Graphical combination of several query results
Display of context
A facility for checking the context of display
Extended dialog
Varying graphical representations
Legend
Label placement
Scale Selection
Subarea for queries

Jiyong Zhang

Januray 25, 2005

Spatial Database: System Architecture

Extensions required to a standard DBMS architecture

16

Representations for the data types of a spatial algebra

Procedures for the atomic operations,
Spatial index structures,
Access operations for spatial index,
Filter and refine techniques
Spatial join algorithms
Cost functions for all these operations (for query optimizer)
Statistics for estimating selectivity of spatial selection and join
Extensions of optimizer to map queries into the specialized query
processing method
Spatial data types & operations within data definition and query
language
User interface extensions to handle graphical representation
Jiyong Zhang

Januray 25, 2005

Spatial Database: System Architecture (contd)

Previous approaches to GIS architecture

Built directly on top of file system

Using a Closed DBMS

1.

2.

17

Jiyong Zhang

Januray 25, 2005

Spatial Database: System Architecture (contd)

Using an Extensible DBMS

There is no difference in principle between:

Current commercial solutions are OR-DBMSs:

18

A standard data type such as a STRING and a spatial data type such
as REGION
Same for operations: concatenating two strings or forming
intersection of two regions
Sort/merge join and spatial join
Query optimization

NCR Teradata Object Relational (TOR)

IBM DB2 (Spatial extenders)
Informix Universal Server (Spatial datablade)
Oracle 8i (spatial cartridges)
Jiyong Zhang

Januray 25, 2005

Temporal Database: Introduction

Most applications of database technology are temporal in

nature:

Financial apps.: portfolio management, accounting & banking

Record-keeping apps.: personnel, medical record and inventory
management
Scheduling apps.: airline, car, hotel reservations and project
management
Scientific apps.: weather monitoring

Definition:

Temporal DBMS manages time-referenced data, and times are

associated with database entities

19

Jiyong Zhang

Januray 25, 2005

Temporal Database: Introduction (contd)

Modeled reality
Database entities
Fact: any logical statement than can meaningfully be
assigned a truth value, i.e., that is either true or false
Valid Time (vt)

Transaction Time (tt)

20

Valid time is the collected times when the fact is true

Possibly spanning the past, present & future
Every fact has a valid time
The time that a fact is current in the database
Maybe associated with any database entity, not only with facts
TT of an entity has a duration: from insertion to deletion
Deletion is pure logical operation
Jiyong Zhang

Januray 25, 2005

10

Temporal Database: Introduction (contd)

Time domain may be discrete or continuous

Typically assume that time domain is finite and discrete in
database
Assume that time is totally ordered

Uniqueness of NOW

The current time is ever-increasing

All activities is happed at the current time
Current time separates the past from the future

NOW <> HERE

21

Time cannot be reused!

A challenge to temporal database management

Jiyong Zhang

Januray 25, 2005

Temporal Database: Modeling

More than 24 extended relational models proposed

Bitemporal Conceptual Data Model (BCDM):

timestamps tuples with sets of (tt, vt) values

UC: until changed

22

Customer C101 rents T1234 on May 2nd for 3 days, and returns it
on 5th
C102 rents T1245 on 5th open-ended, and return it on 8th
C102 rents T1234 on 9th to be returned on 12th. On 10th the rent is
extended to include 13th, but the tape is returned on 16th
Jiyong Zhang

Januray 25, 2005

11

Temporal Database: Modeling (contd)

Graphical Illustration of the Timestamp Values

3. C102 rents T1234 on 9th to
be returned on 12th. On 10th
the rent is extended to
include 13th, but the tape is
returned on 16th

2. C102 rents T1245 on

5th open-ended, and
return it on 8th

1. Customer C101 rents

T1234 on may 2nd for 3
days, and returns it on 5th

23

Jiyong Zhang

Januray 25, 2005

Temporal Database: Modeling (contd)

BCDM pros:

BCDM cons:

24

Simple, while also capturing the temporal aspects of the facts

stored in a database.
Since no two tuples with mutually identical explicit values are
allowed in BCDM relation instance, the full history of a fact is
contained in exactly one tuple.

Bad internal representation and display to users of temporal info

Varying length and voluminous timestamps of tuples are
impractical to manage directly
Timestamp values are hard to comprehend in BCDM format

Jiyong Zhang

Januray 25, 2005

12

Temporal Database: Querying

Temporal queries can be expressed in conventional query

language such as SQL, but with great difficulty
Language design must consider

Time-varying nature of data

Predicates on temporal values
Temporal constructs
Supporting states and/or events
Supporting multiple calendars
Modification of temporal relations
Cursors, views, integrity constraints,
handling now, aggregates, schema versioning, periodic data

Some 40 temporal query languages have been defined

More recent language: TSQL2

Extension to SQL-92

25

Jiyong Zhang

Januray 25, 2005

Temporal Database: DBMS Implementation

Integrated approach: internal modules of a DBMS are

modified or extended to support time-varying data

Layered approach: a software layer interposed between the

user applications and DBMS that converts temporal query
language statements to conventional statements

26

Efficiency

More Realistic for short and medium term

Jiyong Zhang

Januray 25, 2005

13

Spatiotemporal Database: Applications

Three Types of Spatiotemporal Applications

1. Applications may involve objects with continuous motion

2. Applications dealing with discrete changes of and among

objects

Objects shape and their positions may change discretely in time

3. Applications may manage objects integrating continuous

motion as well as changes of shape

27

Navigational systems manage moving objects

Objects change position, but not shape

A storm is modeled as a moving object with changing properties

(e.g., intensity) and shape over time

Jiyong Zhang

Januray 25, 2005

Spatiotemporal database: modeling requirements

Need for representations of objects with position in space and

existence in time
Need to capture the change of position in space over time

28

Continuous change, or discrete change

Need for the definition of attributes of space and organization of

them into layers or fields
Need to capture the change of spatial attributes over time
Need to connect spatial attributes to objects
Need for the representation of spatial relationships among objects in
time
Need for the representation of relationships among spatial attributes
in time
Need to specify spatiotemporal integrity constraints, imposed either
by the user, or by the designer for integrity of the database
Jiyong Zhang

Januray 25, 2005

14

Spatiotemporal database: Querying

Spatial operators(Faria1998)

NORTH(A,B)
AREA(A)
LENGTH(A)
DISJOINT(A,B)

Temporal operators:

BEGIN(A), END(A)
T_BEFORE(A,B)
INTERVAL (start-time, end-time)

29

Jiyong Zhang

Januray 25, 2005

Spatiotemporal database: Querying (contd)

Spatio-temporal Operators

Location-temporal Operator ST_SP(A, T)

Orientation-temporal Operators

The metric of object A at a time value T, ST_AREA(A, T)

Distance between two spatial components A and B at time T:
ST_DISTANCE(A,B,T)

Topologic-temporal Operators

30

Return a boolean value indicating whether there exists specific

relationship between two objects (A and B)
ST_NORTH(A,B) or ST_EAST(A,B), etc

Metric-temporal Operators

Returns the spatial representations of object A valid at time T

Return a boolean value indicating the topologic relationship between

A and B during the time T: ST_DISJOINT(A, B, T)

Jiyong Zhang

Januray 25, 2005

15

Spatiotemporal database: Querying (contd)

Querying examples

2.

31

Jiyong Zhang

Januray 25, 2005

Spatio-Temporal Database Systems Architecture

Standard Relational DBMS with Additional Layer

Combination Architecture

Standard DBMS
Other storage components (such as file system) are used to store
the spatial and temporal data and indexes

Extensible DBMS

32

Implement a spatiotemporal layer on top of a standard relational

database

Object-Relational DBMS

Jiyong Zhang

Januray 25, 2005

16

Spatiotemporal database: open issues

Database size

Legacy systems

Spatial databases contain large amounts of information, temporal

information further increases the database size
Increased difficulty of rapid data retrieval
Using STDB to Replace old systems & data
Building new STIS on existing SIS
Utilization of a data warehouse, enabling several legacy systems
to be incorporated in a data-supply role

Data quality

Errors exist in data gathering

discrete representation of numbers in computer
Temporal dimension further this problem

33

Jiyong Zhang

Januray 25, 2005

Summary

Spatio-Temporal Information systems improve the existing

spatial information system by handling temporal information.
Most existing prototype systems are extensions of existing
spatial systems
Mainly for specific purposes (such as global change research)

34

Its unclear if a generic spatio-temporal information system will be

commonly used

Would be benefit from research in both spatial database and

temporal database

Jiyong Zhang

Januray 25, 2005

17

References
1.

2.

3.

4.

5.

6.

35

Ralf Hartmut Guting, An introduction to Spatial Database

Systems, VLDB Journal 3, 357-399 (1994)
Christian S. Jensen, Introduction to Temporal Database
Research, Temporal Database Management, 2000.
Tamas Abraham and John F. Roddick, Survey of SpaioTemporal Databases, GeoInformatica 3:1, 61-99 (1999)
Dieter Pfoser and Nectaria Tryfona: Requirements,
Definitions and Notations for spatiotemporal Application
environments. ACM GIS98
Nectaria Tryfona and Christian S. Jensen, Conceptual Data
Modeling for Spatiotemporal Applications, GeoInformatica
3:3, 245-268 (1999)
Glaucia Faria, Claudia Bauzer Medeiros, Mario A.
Nascimento, An Extensible Framework for Spatio-Temporal
Database Applications (1998)
Jiyong Zhang

Januray 25, 2005

~END~
Q&A
Januray 25, 2005
Jiyong Zhang([email protected])
Ecole Polytechnique Fdrale de Lausanne (EPFL)
School of Computer and Communication Sciences
CH-1015 Lausanne, Switzerland
36

18