bul 2004
7 Various
ilities are
lient side
thers two
ration of
P.,2003).
rograms
ı may or
strategy,
channel
e shared
geodata
network
nnection
tilizing
listribute
can be
ects, the
ih a link
ablished
iterfaces
second
actually
2's local
manual
imilarly,
ponents
invoked
cols are
IOP) or
hnology
include
A GIS
(that is,
will use
server)
roles to
services
cenario
that the
service
tware
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
The second network strategy actually moves the sofiware
modules (the services) from one node to another. The migration
process uses a HTTP protocol to transfer the required GIS
procedure dynamically into the targeted GIS application. The
downloaded GIS component is stored inside a container, which
binds with the local GIS application. This approach is currently
supported by several technology frameworks, such as Java
applets in a Virtual Machine, or Active X containers. It has not
been widely integrated into GIS environments as yet.
In the first strategy it is hard to move geodata through the
network .Various challenges may occur in this process.
Geodata are valuable resources in decision making so security
issues is influenced in this strategy. Another important problem
is network loading which is failed in this way. Uploading and
downloading high volume geodata in a network fail the
network traffic control. Uploading geodata across the network
and then downloading it is done bye applets or FTP protocol. In
each manner we faced with thick client strategy. Recent
development in object oriented programming make it possible
to produce software components and send them to the client
before the running in the client machine, such as Java classes,
ActiveX components and plug-ins. This comes out the thick
client GIS. Java applet, Netscape plug-ins and Microsoft
ActiveX component technology were involved in the structure.
In the thick client architecture, the client machine does most
processing work locally.
GEOPROCESSING
PROTOCOL
Figure 4: Two types of network connection for Geoprocessing
By combining RPC and sharing geodata objects. thin client
strategy is accessible. In the thin client strategy the client only
have user interface to communicate with the server and display
the results.
In this manner several interactions between client and server is
inevitable especially in the case of GIS processing because of
high memory load and CPU usage, increasing in the number of
the interactions may cause heavy traffic in the network.
Both thin and thick client systems have some advantages and
drawbacks. but they are not the best solution in terms of taking
advantage of network resources.
Combination between shared data and movable software opens
anew paradigm in distributed GIS.
As it is mentioned in the previous section geoprocessing part of
distributed GIS is weak and this solution may solve many
problems such as security, efficiency and network load.
Such movable software across the network are entitled as
agent. Agent technology is a new paradigm in computer
science that next section will discuss about it widely.
OCESSING
711
4. AGENT TECHNOLOGY
Agent concept began by an idea of non-human agencies. It's is
realized by constructing robots. Nowadays it is implemented as
soft robot, living and doing its task within computer world.
An agent is a computer system suited in some environment and
that is capable of autonomous action in this environment in
order to meets its design objectives. There are several
characteristics for an agent, some of them are ideal
characteristics and are far from reality. But some characteristics
like mobility, communication ability, reactivity and inferential
capability can enhance GIS applications in various fields.
4.1 Mobile agent
Mobile agents are a class of agents whose predominant feature
is the ability to transport between nodes on a network or
between nodes across networks. They are the basis upon which
true distributed information management agents can be built.
Mobile (or transportable) agents are a direct extension of the
client/server technology. In the client/server paradigm,
communicating entities have fixed and well-defined roles; a
server offers a set of services and a client makes use of those
services. This model also implies a strict sense of dependency:
clients are dependent upon servers to provide the services that
they require. The communication mechanism that takes place
between a client and a server is through a message passing
protocol since network communication is assumed. However,
message passing has been criticized as being too low level,
requiring programmers to determine network addresses and
synchronization points themselves.
However, a fundamental problem exists with client/server
architectures when considering distributed information system.
If the server does not provide the exact service that the client
requires, for example the server only provides low-level
services, and then the client must make a series of remote calls
to obtain the end service that it requires. This may result in an
overall latency increase and in intermediate information being
transmitted across the network which is wasteful and
inefficient, especially for high volume of spatial data.
Moreover, if servers attempt to address this problem by
introducing more specialized services, then, as the number of
clients grow, the amount of services required per server
becomes unfeasible to support.
The mobile agent paradigm attempts to address the issues that
are raised by the client/server and paradigms. Typical
characteristics of mobile agents are their ability to migrate at
will, autonomy in their actions, a peer-to-peer personality and a
processing and network independence from their original
location.
Mobility is a desirable characteristic in agents for a number of
reasons:
4.1.1 Efficiency: If an agent can move across networks to the
location where resources reside, then network traffic can be
reduced since the agent can preprocess data and decide which
the most important information to transfer is. This is a crucial
aspect when considering users who connect through a low
bandwidth link.
4.1.2 Persistence: Once a mobile agent is launched, it should
not be reliant on the system that launched it and should not be
