Article Preview
Top1. Introduction
This paper sums up the design, development, views and recommendation on Cloud Computing based on a wide variety of articles. Dated back in 2007, Cloud Computing was regarded as a buzz word that came out from marketing teams. However, from that time onwards, different types of definitions, demonstrations, proofs-of-concepts and services have been in the literature and in the market driving different directions and emphasis for the research communities. Buyya et al. (2008) describe that Cloud Computing was regarded as the fifth utility, after water, electricity, gas and petrol, and it could be used as an extensive grid computing. With the use of virtualization, data centers can be managed more effectively by Cloud Computing. He demonstrated his concepts and software to show how Cloud Computing should go ahead. Buyya et al. (2009) also illustrate quality as a service (QoS) as an important factor to Cloud Computing which could affect pricing model and the effectiveness of using Cloud Computing for scientific experiments. If services that have a low QoS, it means there are higher costs involved and also the services are more subject to failures or unsatisfactory performance. Through the development of Buyya’s research project, CloudSim and Aneka have been introduced in different time scale. CloudSim is a Java-based open-source software to simulate the use of Cloud Computing located at different data centers. By using CloudSim, it allows users and developers to understand the electricity consumptions, costs and management of data centers. Services and data can be transferred from one data center to another and factors such as QoS, performance, security and costs can affect the Cloud Computing services as a whole. There are also other different researchers focused on different aspects on Cloud Computing. For example, Foster et al. (2008) define that Cloud Computing was a subset of Grid Computing and drew diagrams to explain the relationship between Clouds, Grids, Clusters and Supercomputers. However, it is incorrect since Clouds are not a full subset of Grids. Grid computing has failed to make it user friendly since the software design and development are heavily centered on the scientists and developers who have significant years of experience in using high-level software. It has struggled to make full impacts to the user community and the consumer market. Additionally, Cloud Computing requires the use of virtualization, which means each virtual machine (VM) can be used as if a physical machine. Original concept of Grid Computing did not use virtualization at all and only up to this paper, Foster et al. (2008) then revised their definition. They argue that problems faced by Grids and Clouds are similar. However it is not entirely true. First, problems in Cloud Computing involved with multi-tenancy, business models and top layers of the service models are not entirely the same or in the similar level as Grid Computing. Second, there are a far more user community based markets for Cloud Computing and user-driven requirements are very different. User requirements in Clouds focus on usability and even lower-end requirements. User requirements in Grids focus on the high-level complexity and often has failed to demonstrate how problems can be resolved in less complicated ways, less costs and less steps involved.
With regard to the use of virtual machines (VMs), it takes much quicker to clone and duplicate VMs than physical servers. It also means that less physical servers are required, since each physical server can host multiple VMs without struggling to cope in performance. Thus, data centers can be consolidated with less space, less machines and as a result, less air-conditioning involved. In other words, it provides cost-savings since less operational costs will be required to host data centers which can be hit by the rise of electricity bills. This offers another benefits of using Cloud Computing, cost-savings, with different papers illustrating their points. For example, Chang et al. (2013 a) demonstrate a framework that can deliver different types of services and achieve cost-savings. Under the framework, there are also other cases studies, details of which will be presented in Section 2 and 3 of this paper. The breakdown of this paper is as follows. Section 2 describes literature review for Cloud Computing and presents each case in details, including supporting papers for each indentified area of significance. Section 3 presents a case for Cloud Computing adoption and recommends the use of framework for Cloud adoption. Section 4 shows topics of Discussion and Section 5 concludes this paper and presents the future work.