Continuous Double Auction Scheduling in Federated Cloud Services

In recent years, researchers have introduced many different mechanisms to improve resource allocation in the cloud. One of these resource allocation methods is market-based resource allocation which exploits different models used in exchanging goods and services. In this research, a two-way auction model is used for allocating cloud resources based on the market model. In the case of federated clouds, as the providers may face a shortage of resources during their operation; therefore, the continuous double auction model is suggested to create a cloud federation environment to support a suitable resource allocation among different providers. In our experiment 1, fixed pricing with Reputation-Aware Continuous Double Auction, Continuous Double Auction, and Market-Driven Continuous Double Auction models will be executed for resource allocation. It shows that both the resource efficiency and the income of the providers are improved in the federated clouds using these models. In experiment 2, with changing the type and number of the requested resources by customers and providers, the proposed federated model is also tested. The results of the experiments show that our proposed model for implementing federated clouds based on the continuous double auction model, in terms of successful allocation rates, resource efficiency and provider revenue, is better than other market-oriented models. 

Keywords:

Resource allocation , Cloud federation , Continuous double auction , Double Auction model