研究服务组合 QoS 模型在云计算中的运用
ABSTRACT
With the rapid development of the Internet, many industries
production capacity gradually move closed to the composition of the
collaborative, intelligent, virtualization, service-oriented, that making the
cloud service has become one of the current research focus. Cloud QoS
model service composition have been studied, including the needs of the
user based on user behavior analysis to obtain their basic individual
needs, reflect the service composition process model for optimal QoS
values obtained and the composite service QoS model be adaptively a
comprehensive evaluation, and an adaptive adjustment algorithm based
on this model.
In this paper, the combination of the service QoS model used in the
cloud carried out relevant research, the work includes the following three
aspects.
1. It proposes personalized user behavior analysis. Since different
knowledge resources are virtualized abstract knowledge into a variety of
services, in order to allow users to choose effective knowledge service,
knowledge service platform for analyzing user access behavior of these
services through a combination of user needs in order to provide
knowledge to the user, so proposed based on the behavior of individual
users of the resource selection policy approach to research on the
different user behavior. With this personalized acts resource selection
policy approach enables users basic needs are met, and in accordance
with the access to the platform of different knowledge resources,
knowledge resources of several weighted average duration and
frequently visited for comparison, the needs of individual users program.
2. To analyse the algorithm performance service portfolio QoS
values. For a large number of services exist in the service library service
composition process leading to slow or unable to effectively meet a
combination of the user's problem, a screening method for optimal and
efficient service QoS values can be drawn on this basis. The method by
finding all Services library screening draw an effective service and filter
out invalid service, all based on the effective collection of services
万方数据
