Reliability-based Formation of Cloud Federations Using Game Theory

Alam, A B M Bodrul; Halabi, Talal; Haque, Anwar; Zulkernine, Mohammad

doi:10.1109/globecom42002.2020.9322177

Cited by 2 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the simulation, Poisson distribution is used for generating requests and occurring failures. Due to the popularity and applicability of Poisson distribution in modeling request and failure generation in cloud, it is the most appropriate distribution [35], [36], [37]. At first, the initialization of all required parameters for simulation is set to obtain the necessary data.…”

Section: Simulatormentioning

confidence: 99%

A Resource Allocation Model Based on Trust Evaluation in Multi-Cloud Environments

2021

View full text Add to dashboard Cite

Allocating and managing resources while considering the quality of service is considered a fundamental and complex research problem in a cloud environment. An optimal resource allocation optimizes several parameters such as optimizing cost and resource utilization or maximizing any quality parameters. However, to ensure better customer service, Cloud Service Providers (CSPs) should consider most of the quality attributes while allocating resources to the cloud infrastructures. Existing research does not evaluate trust as a quantitative attribute, thus a trade-off between trust and performance in resource allocation is also absent in the research area. We propose a model to consider both trust and delay in this paper. The trust of a CSP is quantitatively estimated through some attributes and metrics. Availability, reliability, data integrity, and efficiency are considered to estimate the trust. The objective is to maximize the trust of the allocation while minimizing the communication delay. The proposed joint optimization model combines the previous credentials of the CSPs and the present resource constraints. To solve the problem heuristically, a genetic algorithm is applied. The model uses a number of parameters that provide the flexibility to adapt several service requirements. The effectiveness and applicability of the proposed approach are demonstrated through experiments. The results ensure that the effectiveness in the estimation of trust evaluation for different CSPs with the proposed attributes. Moreover, integrating trust in the resource allocation model allocates appropriate resources while enhancing the trust and reducing the communication delay in the overall allocation.

show abstract

Section: Simulatormentioning

confidence: 99%