| Rajkumar Buyya1,2, Rajiv Ranjan3, Rodrigo N. Calheiros1
The paper "InterCloud: Utility-Oriented Federation of Cloud Computing Environments for Scaling of Application Services" by Rajkumar Buyya, Rajiv Ranjan, and Rodrigo N. Calheiros discusses the challenges and solutions for dynamically coordinating load distribution among different cloud-based data centers to achieve optimal QoS levels. The authors propose the creation of a federated cloud computing environment (InterCloud) that facilitates just-in-time, opportunistic, and scalable provisioning of application services, adapting to variable workload, resource, and network conditions. The goal is to support dynamic expansion or contraction of capabilities (VMs, services, storage, and databases) to handle sudden variations in service demands.
The paper outlines the vision, challenges, and architectural elements of InterCloud, emphasizing the need for automatic load coordination and service distribution based on changing loads. It highlights the importance of predicting geographic distribution of users and the limitations of current cloud computing systems in meeting QoS expectations across multiple geographical locations.
The authors present a detailed system architecture consisting of a Cloud Coordinator, a Cloud Broker, and a Cloud Exchange. The Cloud Coordinator manages domain-specific enterprise clouds and their membership in the federation, while the Cloud Broker acts as a mediator between service consumers and Cloud Coordinators. The Cloud Exchange acts as a market maker, facilitating the trading of cloud services based on competitive economic models.
The paper also includes performance evaluation results using the CloudSim toolkit, demonstrating that the federated cloud computing model offers significant performance gains in terms of response time and cost savings under dynamic workload scenarios. The experiments show that federated infrastructure reduces average turnaround time by more than 50% and improves makespan by 20%.
Finally, the authors discuss future directions, including developing comprehensive model-driven approaches for provisioning and delivering services in federated environments, and addressing energy efficiency in data centers to support Green Cloud computing.The paper "InterCloud: Utility-Oriented Federation of Cloud Computing Environments for Scaling of Application Services" by Rajkumar Buyya, Rajiv Ranjan, and Rodrigo N. Calheiros discusses the challenges and solutions for dynamically coordinating load distribution among different cloud-based data centers to achieve optimal QoS levels. The authors propose the creation of a federated cloud computing environment (InterCloud) that facilitates just-in-time, opportunistic, and scalable provisioning of application services, adapting to variable workload, resource, and network conditions. The goal is to support dynamic expansion or contraction of capabilities (VMs, services, storage, and databases) to handle sudden variations in service demands.
The paper outlines the vision, challenges, and architectural elements of InterCloud, emphasizing the need for automatic load coordination and service distribution based on changing loads. It highlights the importance of predicting geographic distribution of users and the limitations of current cloud computing systems in meeting QoS expectations across multiple geographical locations.
The authors present a detailed system architecture consisting of a Cloud Coordinator, a Cloud Broker, and a Cloud Exchange. The Cloud Coordinator manages domain-specific enterprise clouds and their membership in the federation, while the Cloud Broker acts as a mediator between service consumers and Cloud Coordinators. The Cloud Exchange acts as a market maker, facilitating the trading of cloud services based on competitive economic models.
The paper also includes performance evaluation results using the CloudSim toolkit, demonstrating that the federated cloud computing model offers significant performance gains in terms of response time and cost savings under dynamic workload scenarios. The experiments show that federated infrastructure reduces average turnaround time by more than 50% and improves makespan by 20%.
Finally, the authors discuss future directions, including developing comprehensive model-driven approaches for provisioning and delivering services in federated environments, and addressing energy efficiency in data centers to support Green Cloud computing.