الفهرس 
Introduction to Cloud ComputingOnly 14 pages are availabe for public view
 |  | from | 132 |  |  |
| | | from | 132 | | |
Abstract
The emergence of cloud computing over the past few years is potentially one of the major advances in the history of computing. The evolution of cloud computing represents a fundamental change in the way Information Technology (IT) services are developed, deployed, scaled, updated, and paid for. Cloud computing service models can be classified into three main categories; Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS), and Software-as-a-Service (SaaS). One of the major challenges in cloud computing environments is cloud bursting. This term refers to how to integrate public cloud resources along with internal IT infrastructure (private cloud or internal data center). Hence, cloud resources are provisioned when local resources are insufficient to meet their load spikes. In this case, extra workloads will be offloaded to public cloud on demand while private cloud owners are charged on a Pay-As-You-Go (PAYG) basis. Several factors should be determined accurately before and during the bursting process such as the type of resources to be gained from public cloud, the amount of resources needed to cover the needs of cloud clients, and the quality of service of the private cloud owners … etc. However, several hurdles stand in the way of deploying cloud bursting such as private and public clouds are located in different administration domains. Additionally, IaaS services have different technical requirements and may require different Application Programming Interfaces (APIs) for cloud bursting.
There is no comprehensive approach for treating the issues of cloud bursting. Most researches try to develop one of the aspects of bursting process, either by suggesting different frameworks for utilization to solve problems such as administration or security, or through reusing certain techniques to enhance the performance and throughput of the process by focusing on a specific metric such as resource balance or responsiveness. Hence, several cloud bursting issues are left unsolved.
In this thesis, a new cloud bursting framework named “Platypus” is introduced mainly to resolve some of the unresolved issues in cloud bursting. Platypus framework achieves the utter ratio of applying the pre-copy technique and supporting the automation procedures of implementing the cloud bursting approach. In addition, experimental studies based on real values from Amazon Web Services (AWS) has been conducted for applying this new framework, and the framework succeeds in answering the question ”which application will be pre-copied and consequently bursted to the cloud?” Moreover, it achieves the trade-off between the time saving and the cost reduction.
Key words: cloud computing; hybrid cloud; cloud bursting; pre-copy.