vice
patial
QoS)
nents
ssary
vered
QoS-
ution
or GI
rvice
4WS
dane,
users
1s GI
QoS)
tion-
ingly
a re-
1cern
x GI
le GI
nains
10del
sport
uted
. 2001,
high
T ac-
ands
QoS
| Ser-
ta in
Users
osed
; that
idual
leris-
hore,
time,
ired.
rvice
and
| from
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
a community to participate in a business processes based
on their non-functional quality attributes. Once selected,
the services can then be choreographed in their correct
sequence and their execution managed by a QoS-aware
workflow management system to ensure adherence to QoS
specifications.
Further, a sustainable QoS-aware service infrastructure is
predicated on dynamic and collaborative GI marketplaces
in which agile and effective Gl enterprises are service nodes.
How existing enterprises will evolve to be effective part-
ners in the service infrastructure still remains very much
an open issue.
The paper defines an architecture for QoS-aware discov-
ery and composition of GI services and an extensible QoS
model for GI services. The quality model defines key non-
functional quality characteristics of GI services and service
chains. The paper further defines an enterprise integration
framework for GI enterprises towards a dynamic GI market
based on the geographic information service infrastructure.
The framework identifies the fundamental building blocks
for designing, developing and operating highly integrated
Gl enterprises as effective service nodes on the service in-
frastructure.
2 THE GI SERVICE INFRASTRUCTURE
Over recent years, the Gl service infrastructure has emerged
(Alameh, 2001, Radwan and Morales, 2002) presenting
an evolutionary framework for delivering geospatial solu-
tions, products and services that meet constantly changing
user requirements . It extends the capabilities of the GDI
with services enabling delivery of value-added information
products beyond raw spatial data.
The GI service infrastructure is made up of disparate nodes
that collaborate to deliver complex services via the Web,
the nodes being enterprises with business goals. Through
business-to-business (B2B) standards like Rosettanet, EDI
and BPEL4WS, the nodes can dynamically discover each
other, negotiate innovative business propositions, and col-
laboratively execute common business processes to address
a market opportunity and meet strategic mission. The ser-
vice infrastructure therefore sets the foundation for GI e-
commerce and business integration.
3 QOS-AWARE SERVICE CHAINING
Quality of service (QoS) is widely used in various disci-
plines with equally diverse meanings. Table 1 presents
some definitions from literature. More abstract definitions
like ‘quality is what the customer says it is’ or ‘quality is
customer satisfaction’ are not uncommon. The common
theme in all these definitions however, is the abstract no-
tion of user satisfaction, which clearly has many perspec-
tives.
In this paper we focus on both qualitative and quantita-
tive measures of quality of service and section 4 defines
213
User perceived performance
The collective effect of service per-
formances which determine the de-
grec of satisfaction of the user of a
service
The degree of conformance of the
service delivered to a user by a
provider with an agreement be-
tween them
A set of quality requirements on the
collective behavior of one or more
objects and may be specified in a
contract or measured and reported
after an event
QoS is the ability to provide re-
source assurance and service differ-
entiation
Quality is the totality of the char-
acteristics of a product (or service)
that bear on its ability to satisfy
stated or implied needs
Table 1: Some definitions of QoS
(Franken, 1996)
(ITU-T, 1994)
(P806-GI, 1999)
(ITLIASO, 1995)
(Wang, 2001)
(ISO, 2000)
some them. The required QoS at any time will thus be
constraints on any of the measures or specified combina-
tions of these measures. It suffices to note that the list in
section 4 in not in any way exhaustive.
The notion of quality presented in this paper departs from
classical spatial data quality. This is because while high
quality spatial data is necessary, its not sufficient for QoS
in a web services environment. QoS comprises systemic
and business process quality criteria like performance, cost,
reliability etc., over and above spatial data quality.
| Service Discovery Mechanism I 3
| |
i rete SEA
|
pes | Î 1:4 E.
| | | p \
| Client | | Service Chain Generator H KR Ze A
i 1 x :
Si | | n =n]
| p 3 it» Service |
{ Service Orchestration & | | | . |
{ Execution Management Engine | ! | Registry |
| Geographic Information Service Instances
|
Figure 1: Architecture for QoS-Aware Service Chaining
The concept os QoS-aware service discovery and composi-
tion is particularly relevant to translucent and opaque chain-
ing (ISO/TC211, 2002) where naive users are at play. In
an operational environment, users need only to specify the
services they need and corresponding QoS requirements
and should not be bothered with the complex issues of
defining the service chain, discovering the suitable services,
orchestrating them and monitoring their execution to en-
sure achievement of QoS targets. These issues should be
transparent to the user.