InterCloud: Utility-Oriented Federation of Cloud Computing Environments for Scaling of Application Services (1003.3920v1)

Abstract: Cloud computing providers have setup several data centers at different geographical locations over the Internet in order to optimally serve needs of their customers around the world. However, existing systems do not support mechanisms and policies for dynamically coordinating load distribution among different Cloud-based data centers in order to determine optimal location for hosting application services to achieve reasonable QoS levels. Further, the Cloud computing providers are unable to predict geographic distribution of users consuming their services, hence the load coordination must happen automatically, and distribution of services must change in response to changes in the load. To counter this problem, we advocate creation of federated Cloud computing environment (InterCloud) that facilitates just-in-time, opportunistic, and scalable provisioning of application services, consistently achieving QoS targets under variable workload, resource and network conditions. The overall goal is to create a computing environment that supports dynamic expansion or contraction of capabilities (VMs, services, storage, and database) for handling sudden variations in service demands. This paper presents vision, challenges, and architectural elements of InterCloud for utility-oriented federation of Cloud computing environments. The proposed InterCloud environment supports scaling of applications across multiple vendor clouds. We have validated our approach by conducting a set of rigorous performance evaluation study using the CloudSim toolkit. The results demonstrate that federated Cloud computing model has immense potential as it offers significant performance gains as regards to response time and cost saving under dynamic workload scenarios.

• The paper introduces a federated cloud model that dynamically allocates resources to deliver scalable application services.
• It details key components such as the Cloud Exchange, Coordinator, and Broker to optimize load distribution and meet QoS targets.
• Experimental results using CloudSim demonstrate over 50% reductions in task turnaround times and significant cost savings compared to isolated setups.

InterCloud: Utility-Oriented Federation of Cloud Computing Environments for Scaling of Application Services

The paper "InterCloud: Utility-Oriented Federation of Cloud Computing Environments for Scaling of Application Services" by Rajkumar Buyya, Rajiv Ranjan, and Rodrigo N. Calheiros addresses the creation of a federated cloud computing environment named InterCloud. This environment aims to offer just-in-time, scalable, and opportunistic provisioning of application services across multiple cloud providers, ensuring the achievement of Quality of Service (QoS) targets despite variable workloads and resource conditions.

Vision and Challenges in Cloud Federation

The paper posits that current cloud platforms, which include prominent players like Amazon AWS, Google App Engine, and Microsoft Azure, lack robust mechanisms for dynamic load distribution across geographically distributed data centers. Cloud providers typically expect their customers to specify hosting preferences, failing to automate the selection and dynamic allocation of resources based on real-time demand fluctuations. This inability limits the scalability and QoS delivery, especially to a globally distributed user base.

Architectural Elements

The proposed InterCloud architecture involves several key components:

1. Cloud Exchange (CEx): Acts as a market maker, facilitating resource trading based on SLAs and competitive economic models like auctions. It plays a critical role in dynamic service discovery and resource allocation.
2. Cloud Coordinator (CC): Manages resource provisioning within an individual cloud domain, interfacing with the CEx for resource offers and ensuring compliance with energy management and cost goals. Components within the CC include the Scheduling & Allocation module, Market & Policy Engine, Application Composition Engine, Virtualization Manager, and Sensor Infrastructure.
3. Cloud Broker (CB): Operates on behalf of users, negotiating with cloud coordinators through CEx to meet application-specific QoS targets. The CB determines optimal resource allocation strategies and monitors cloud services dynamically.

Experimental Validation

The research validates the proposed federation model using the CloudSim toolkit. The experimental setup juxtaposes a federated cloud environment against isolated cloud setups, showing significant improvements in both performance and cost-efficiency for dynamic workloads.

Key Findings:

• Performance Gains: Federated clouds reduce average task turn-around time by over 50%, with a 20% improvement in makespan, indicating enhanced resource utilization and workload distribution efficiency.
• Cost Efficiency: Dynamic leasing of resources from public clouds during peak demand significantly reduces processing costs as compared to maintaining over-provisioned private cloud infrastructure.

Implications and Future Directions

The implications of implementing an InterCloud federation are profound, enabling cloud service providers to offer superior scalability and reliability while maintaining cost-effectiveness. For practitioners, the federation model facilitates seamless integration of multiple cloud platforms, enhancing the resilience and performance of cloud-hosted applications.

The paper outlines several promising future avenues for research:

1. Advanced Provisioning Models: Developing sophisticated analytical and statistical models to predict application behavior and resource requirements accurately.
2. Energy-efficient Resource Management: Creating algorithms that optimize resource allocation while minimizing energy consumption, critical for sustainable cloud computing.
3. Robust Monitoring Infrastructure: Implementing scalable, decentralized monitoring techniques to ensure real-time dynamic adjustments in resource provisioning across federated clouds.

Conclusion

The research presents a comprehensive and validated framework for federating cloud services across multiple domains, addressing critical scalability and QoS challenges posed by existing cloud infrastructures. InterCloud represents a significant advancement towards realizing a highly adaptive, energy-efficient, and cost-effective cloud ecosystem capable of meeting the dynamic needs of a global user base. Future work will focus on refining provisioning models, improving energy efficiency, and developing robust monitoring techniques to enhance the efficacy of the federated cloud model.