Although these "clouds" represent the natural evolution of traditional data centers, they are distinguished from the latter in the modes of supplying resources, whether they are computing, storage or application, typically exposed as web services. The costs of these resources are charged to the user based on the actual use, allowing substantial savings, even by several orders of magnitude.

Definitions

There are several definitions attributed to the term "Cloud Computing":

- Buyya et al. so define: “Cloud is a parallel and distributed computing system consisting of a collection of inter-connected and virtualised computers that are dynamically provisioned and presented as one or more unified computing resources based on service-level agreements (SLA) established through negotiation between the service provider and consumers.”

- Vaquero et al. assert: “clouds are a large pool of easily usable and accessible virtualized resources (such as hardware, development platforms and/or services). These resources can be dynamically reconfigured to adjust to a variable load (scale), allowing also for an optimum resource utilization. This pool of resources is typically exploited by a pay-per-use model in which guarantees are offered by the Infrastructure Provider by means of customized Service Level Agreements.”

- A report from the University of California Berkeley summarizes the main characteristics of cloud computing as follows:

"(1) the illusion of infinite computing resources;

(2) the elimination of an up-front commitment by cloud users;

(3) the ability to pay for use . . . as needed . . ."

Metrics

Until today few studies have been carried out about the control of effort and quality of Cloud Computing applications, besides the simple measures of software modules installed in the cloud.

The research carried out by the Agilab Group in this field, uses an advanced test bed:

during 2012, in Genoa labs at DITEN and in Cagliari at the Department of Electrical and Electronic Department of the University, the first computational, storage and network systems have been installed  to realize the core of the federated cloud, that will be partially utilized for the experiments of this research.

Thanks to the extreme flexibility and reliability of these systems and sophisticated resource partitioning, an important part of the plant already hosts a set of virtualized services for public use.

The first experiments will address the profiling of the virtualization hosts and the subsequent availability of a large number of measurements will allow to correlate the results of the tests performed in a controlled environment, with other tests carried out in public cloud resources, both academic and commercial:
we will then deduce some of the key factors for determining the quality of the service, considering for example the degree of over-committing resources applied by the provider.