Prioritizing Service Based on Multi-Tier Service Approximation Technique for Dynamic Selection and Composition of Cloud Services in Multi Tier Clouds

The increasing amount of information and the growth of information technology paved the way for internet users to perform their desired task without spending more on the resources using cloud computing. The cloud provides various services at different level of architecture like service layer, platform layer and database layer. To access these services the users has to be registered and could access the desired information through appropriate service provided by the service provider. There will be more than one service available to perform specific task and the service itself may require a different level service. The throughput of the environment is highly depending on how many service requests has been fulfilled on any particular time. Even though there exist number of service the service selection makes the difference and there are many approaches have been discussed earlier to perform service selection, but suffer with the problem of unavailability, poor completion status, time complexity and latency. To overcome these issues, we propose a multi tier service based service prioritization scheme to perform service selection. The selected services will be bundled to produce a composite service which will be given to the user request to complete the task. The service approximation technique computes service depthless measure and service throughput measure. Based on these two, we compute the service trust weight to prioritize the services. A top weighted service will be selected to fulfill the service request of the user and reduces the latency and time complexity of service selection. The proposed approach increases the throughput of overall environment and reduces the time complexity.