In order to provide highly available systems, cluster systems are used in information technology. Cluster systems are a larger number of networked computers that increase computing capacity or availability relative to a single computer.9 In cluster systems, key components of the system are kept redundant. In the case of malfunctions, various programs switch over, resulting in only short downtimes or the loss of an open transaction. To ensure the smooth and efficient operation of systems, dotSource GmbH10 offers a comprehensive hosting and support service. Systems are hosted on the existing cluster system in an external data center. The cluster system of dotSource GmbH, see Appendix A, serves as an analogy model in relation to Cloud Computing Services. The cluster system currently deployed consists of several application servers, two database servers, two load sharing servers, a network attached storage (NAS) system, and a backup server. The physical servers are not provided by the company dotSource for financial reasons, but are rented in the data center of “Manitu” 11. The operator of the data center provides in this context only the physical devices and is responsible for support inquiries. Server, done by the administrator of dotSource GmbH.

definition

For the term cloud computing there is still no standardized, uniform definition. The National Institute of Standards and Technology describes cloud computing as follows.

Cloud computing is a model that allows the user to access a common pool of configurable computing resources through on-demand network access. The resources are scalable and can be deployed and published quickly. The cost of management thus falls to a negligible 12

The basic concept of cloud computing uses virtualization to provide resources of various kinds electronically as services. Virtualization is the foundation of cloud architecture. The pooled resources from the resource pool can solve different requirements as needed. The resources are dynamically adjusted so that they are used to fulfill the task.

Cloud architecture

Viewing the cloud architecture differs from a technical point of view in the various objectives, architectures and services. There are generally three categories that differ in their components and functions. See Figure 2: Cloud Architecture.

SaaS – Software as a Service

Software as a Service includes applications that can be run from the cloud. The user is left with a local installation due to the cloud operation, because the required resources are obtained via the Internet.13 Figure 3: SaaS software as a service

The provider of the cloud software also provides the storage space for accumulating data for the software. This results in a high responsibility for availability and data security. There are generally two billing models for this model. The user pays either a fixed amount per time unit or an amount which corresponds to the degree of use. A well-known application comes from salesforce.com.14 The application offers business solutions to customers as a service over the Internet. Among others, this includes the CRM solution.

PaaS – Platform as a Service

The layer platform as a service is not aimed, as SaaS to end customers, but is aimed at developers. The developers are offered a development environment and an execution environment at PaaS. To perform the hardware and installation or administration of e.g. The developer does not have to worry about the operating system. With these environments, the developer is allowed to develop and execute own programs in a specific programming language. In this area, the Google App Engine17 is a well-known representative. The App Engine only allows the user to develop Python or Java based programs and operate them on Google’s servers.

Detailed description of cloud service providers

There are many providers of cloud computing on the market, see overview tables on cloud architecture. The market offer, however, is still very manageable. Most providers come from the American area after analysis by the author.

If German companies opt for such a cloud provider, it automatically comes to the problem that important company data is stored outside the country on foreign storage resources. The German providers such. Deutsche Telekom, Nionex and Pironet NDH Datacenter can thus exploit their location advantage. The major providers such as Amazon, Google, Salesforce or IBM are anxious to secure their market share. For these reasons, Amazon already set up cloud services in Europe in late 2008 to secure its market share. The market nevertheless looks very clear.

Increasing competition and screening of smaller suppliers will lead to an innovative and cost-effective market, according to the Experton Group. Users should check whether the offered IaaS providers, the desired scalability and the billing models correspond to what vendors offer as marketing promises.

For some cloud services, the following chapter will give an overview and their functionality. The selection fell on these offerings because they enjoy a high presence, as already explored by the Cloud Vendor Benchmark. At this point, it should also be mentioned that the cloud computing field is by far not just the following providers.