查看文章

We present and contrast two solutions for the provision of a distributed file processing system using Amazon’s AWS Cloud. Each system was implemented by independent teams of MSc ITEC students, with the resulting systems providing much functionality which is analogous, but with markedly differing implementation strategies and focus. Cloud Computing is a neologism that describes an internet based service that provides computing resources on-demand, often in a pay-as-you-go model akin to a public utility. It aims to provide an abstraction from the complexity of the underlying distributed infrastructure, in order to offer access to a seemingly endless pool of computational resources. The user should require little or no interaction with the cloud provider, with the service being able to expand and contract on demand from the user, taking advantage of the metered service. Furthermore most cloud services provide additional services to ease development in a distributed environment, such as queue and database facilities. The two solutions attempt to make best use of the tenets of Cloud Computing;“on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service”, and we first explore the architecture of each system are explored in detail, followed by a critical comparison and analysis of the strengths and limitations of each implementation, in addition to suggestions for future possible improvements and evolution.