Research Of Service Assurance Mechanism Based On Cloud Service Bus

With the advent of the era of cloud computing, many cloud platforms of distinctcategories come into beings in recent years. Since the cloud is a broad collection ofservices, different types of cloud computing services commonly are referred toInfrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service(SaaS) according to the service model. PaaS providers support the development anddeployment of applications. However, some important issues still exist in major PaaSproviders such as Google App Engine, Force.com, Heroku, etc. Applications can’t bereused within these PaaS-based cloud platforms. Besides, PaaS providers almost do notsupport Service Level Agreement (SLA) as well as Quality of Service (QoS) guarantees.In order to solve these issues mentioned above, a service assurance system based oncloud service bus is developed in this paper. The application is released in the form of aweb service, by which the applications’ functionalities are successfully fulfilled by callingthe web service. Moreover, the reusability of web services have also been efficientlyenhanced through shared services. Meanwhile, the developed system also offerson-demand QoS adaptability in SLA. Cloud service bus is a distributed service bus basedon P2P architecture, which is effectively used to monitor cloud services. On the other hand,service management module guarantees the high availability and scalability based oncloud service bus. It has efficiently improved the resource utilization, which has notimpacted the quality of service.Moreover, the experimental results show that service assurance system hassuccessfully fulfilled the function of service assurance by functional test. It resolves thesingle point of failure and also guarantees the quality of service based on SLA.Performance analysis shows that the average response time has been cut down by50%aswell as the throughput being doubled with2service instances than that of Axis.Performance is improved efficiently with the increasing number of service instances.