of the Internet. The move from best effort to some guaranteed level of service not only
applies to IP traffic, it applies to the components that are being used to deliver services over
the Web and even the services themselves. The mechanism that allows guarantees to be
implemented is what's termed, Quality of Service . Quality of Service (QoS) has a number
of different meanings depending on the context it is talked about. In the context of networks,
it can be define as the ability of a network element (e.g. an application host or router) to have
some level of assurance that its traffic and service requirements can be satisfied [2]. In the
simplest sense, Quality of Service (QoS) means providing consistent, predictable data
delivery service. In other words, satisfying customer application requirements [3]. This latter
definition is perhaps, a more suitable general definition, as QoS can be discussed at a
number of different architectural levels including network, application and end user QoS.
For example, implementing QoS so that specific user(s) can receive a better that best effort
service from a Webserver has been researched [4].
The formal specification of the QoS guarantees that any of these components implement is
defined with what is termed, a Service level Agreement. A Service Level Agreement (SLA)
specifies the expected performance and availability of a service component or an overall
service that may involve multiple components, or services. However QoS assurances are
only as good as the weakest link in the chain between sender and receiver. If any single
component to which a service depends on fails, or has performance issues, the entire service
is impacted. For example, a Web based Customer Resource Management (CRM) system
may be a collection of independent services such as network, Webserver, and database
service(s), but from a customer s point of view this is just one service. The entire CRM
service may be deemed unavailable if any one of the underlying services (such as the
database) is unavailable.
The dependencies of Web delivered services on multiple components makes it apparent that
a composite view of QoS and availability for all service components is needed when services
are chained and delivered as a single service. However, specific components in a service
delivery chain may not managed by the Service Provider and hence this information may not
always be available. For example, the network that the end user is accessing the service from