Abstract
With the development of large-scale data centers, how to reduce the energy consumption of data centers has becoming one of the most important issues in the cloud computing. To provide the better solution to resource allocation across data centers, one resource allocation scheme based on the Priority Combinatorial Double Auction (PCDA) is proposed. In the PCDA, the auction process of cloud resources includes three phases: cloud user brokers quote, cloud resource providers bidding, and the resource administrate negotiation the bidding. Through the priority of users and the urgency of tasks, PCDA would estimate the SLA (Service Level Agreement) violations for each task during the auction process, and compute the profit of cloud providers. At the same time, multiple transactions are allowed per round. Cloud resources would be allocated reasonably based on the user levels. Simulation results demonstrate that the proposed PCDA can guarantee the bidding success rate. Compared with the traditional combinational double auction (CDA) - one auction, one transaction, the proposed PCDA can reduce the energy consumption by 35.00% during the auction period, and increase the profit of cloud providers by about 38.84%.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call