Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Recurso digital |
| Lenguaje: | |
| Publicado: |
Zenodo
2025
|
| Acceso en línea: | https://doi.org/10.5281/zenodo.15269103 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
Tabla de Contenidos:
- <p>With the rise of<a href="https://ijetrm.com/issues/files/Apr-2025-23-1745412758-APR48.pdf" target="_blank" rel="noopener"> cloud computing</a>, an increasing number of applications are being deployed in cloud environments.<br>A key characteristic of cloud computing is its pay-as-you-go model. Despite this flexibility, many users end up<br>paying more than what they actually use due to the standard one-hour billing interval. Additionally, cloud service<br>providers often offer discounts for long-term commitments, which are inaccessible to short-term users with<br>minimal computing needs. To address these issues and help users reduce expenses, a new entity known as a cloud<br>broker is introduced. This broker acts as a mediator between cloud service providers and end users. It purchases<br>reserved virtual machines (VMs) from providers at a discounted rate and resells them to users on-demand at lower<br>prices than what providers usually charge. Cloud brokers also implement shorter billing cycles, further helping<br>users to minimize costs. Besides benefiting users, cloud brokers can also generate profits by capitalizing on the<br>price difference between reserved and on-demand VMs. This paper explores strategies to effectively configure a<br>cloud broker and determine optimal VM pricing to maximize profits while simultaneously reducing costs for<br>users. Several factors influence the broker’s profitability, including user demand patterns, acquisition and selling<br>prices of VMs, and the broker’s operational scale. These variables are interdependent, adding complexity to the<br>profit analysis. We begin by examining these factors comprehensively and then formulate a profit maximization<br>problem, which involves determining the best multi-server setup and VM pricing scheme. To solve this problem,<br>we introduce a heuristic approach that combines partial derivatives with bisection search techniques. The resulting<br>near-optimal solutions can assist brokers in making informed decisions about resource configuration and pricing.<br>Experimental results and comparisons demonstrate that cloud brokers can substantially reduce costs for users<br>while maintaining profitability</p>