Abstract
Double auctions are considered to be effective price-scheduling mechanisms to resolve cloud resource allocation and service pricing problems. Most of the classical double auction models use price-based mechanisms in which determination of the winner is based on the prices offered by the agents in the market. In cloud ecosystems, the services offered by cloud service providers are inherently time-constrained and if they are not sold, the allocated resources for the unsold services are wasted. Furthermore, cloud service users have time constraints to complete their tasks, otherwise, they would not need to request these services. These features, perishability and time-criticality, have not received much attention in most classical double auction models. In this paper, we propose a cloud priority-based dynamic online double auction mechanism (PB-DODAM), which is aligned with the dynamic nature of cloud supply and demand and the agents’ time constraints. In PB-DODAM, a heuristic algorithm which prioritizes the agents’ asks and bids based on their overall condition and time constraints for resource allocation and price-scheduling mechanisms is proposed. The proposed mechanism drastically increases resource allocation and traders’ profits in both low-risk and high-risk market conditions by raising the matching rate. Moreover, the proposed mechanism calculates the precise defer time to wait for any urgent or high-priority request without sacrificing the achieved performance in resource allocation and traders’ profits. Based on experimental results in different scenarios, the proposed mechanism outperforms the classical price-based online double auctions in terms of resource allocation efficiency and traders’ profits while fulfilling the double auction’s truthfulness pillar.
Highlights
The cloud ecosystem is a business model that needs an appropriate pricing mechanism to satisfy service providers, as well as service users to survive and grow in the current competitive markets [1]
Based on the suggested model, we propose a mechanism that improves the successful-trade rate by increasing the matching rate factor and the overall utility
Experiment setup In our simulation environment, there are 10 service providers as sellers and 10 service users as buyers that participate in the market in 5 separate time-slots, and every time-slot lasts for 30 min
Summary
The cloud ecosystem is a business model that needs an appropriate pricing mechanism to satisfy service providers, as well as service users to survive and grow in the current competitive markets [1]. Since in the on-demand pricing model, customers of cloud computing resources have full control over their operational costs, and they can start and end the use of resources according to their needs, the on-demand pricing model is desirable for them [2, 3]. Calculating the amount needed to invest in infrastructure and the right number of support staff, and scheduling the VMs efficiently, require sufficient knowledge of the needs of users in future time windows, which on-demand pricing models cannot provide
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
