Credibility based multi attribute Combinative double auction for resource Allocation in cloud computing

Abstract

Cloud computing is a growing technology where the resources are provided as a service on demand basis. The services offered are Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS), Network as a Service (NaaS) etc. Based on the requests or the workloads received from the customer side, the resources are fairly allocated to the cloud customers in order to complete their jobs in time. As there exists huge volume of resources in cloud computing, plenty of workloads from various customers are submitted to the cloud workload analyzer. Identifying, analyzing and clustering of the huge volume of workloads for resource allocationis a complex task in the cloud computing environment.It is also difficult to match the workloads and resources based on the expectations of customers and the providers. This thesis explores three important aspects in finding the best workload-resources pairs for resource allocation in cloud environment. newlineIn this thesis, an Extended Cloud Dempster-Shafer Theory (ECDST) based clustering model is proposed in the first phase, for identifying, analyzing and clustering the workloads efficiently. The experimental result demonstrates that the proposed Extended Cloud Dempster-Shafer Theory (ECDST) based clustering modelperforms clustering of the workloads efficiently and accurately by comparing its performance with existingQoS attribute s weight based clustering model. newlineIn Cloud computing, there exista greater number of heterogeneous resources by the cloud service providers andlarge number of workloads are submitted by the customers simultaneously, it is difficult to match the suitableworkloads with the resources based on the expectations of customers and providers. Therefore, in the second phase newline