Heterogeneous Cloud Resource Management: A Comparative Study of Nash Auction and Energy-Aware Nash Auction Mechanisms

Heterogeneity within the cloud infrastructure has also been among the factors responsible for the problem of effective resource management. Existing allocation algorithms are unable to balance the conflicting interests of cloud providers (revenue maximisation) and consumers (Quality of Service). Two game-theoretic auction mechanisms of resource management are introduced and compared in this paper: the traditional Nash Auction and an Energy-Aware Nash Auction. The primary objective is to compare their performance across resource utilisation, user satisfaction, provider revenues, and energy consumption. A simulated multi-cloud facility constructed with the CloudSim toolkit is used throughout the research. Synthetic data comprising 477 unique data samples, each creating user virtual machine requests along with physical machine specifications, was employed to commence the simulation. The Energy-Aware Nash Auction mechanism proposed adds an energy-consumption penalty to the provider's utility function to incentivise deployment on low-energy resources. Findings, derived from the Python, Pandas, and Matplotlib libraries, reveal that while the fundamental Nash Auction maximises providers' revenue, the Energy-Aware Nash Auction achieves substantial energy savings in total energy consumption with minimal effects on revenue and user satisfaction, and is a deployable model for green or eco-friendly cloud data centres.