The deployment of a public cloud computing system is characterized on the one hand by the public availability of the cloud service offering and on the other hand by the public network that is used to communicate with the cloud service. The cloud services and cloud resources are procured from very large resource pools that are shared by all end users. These IT factories, which tend to be speciﬁcally built for running cloud computing systems, provision the resources precisely according to required quantities. By optimizing operation, support, and maintenance, the cloud provider can achieve signiﬁcant economies of scale, leading to low prices for cloud resources. In addition, public cloud portfolios employ techniques for resource optimization; however, these are transparent for end users and represent a potential threat to the security of the system. If a cloud provider runs several datacenters, for instance, resources can be assigned in such a way that the load is uniformly distributed between all centers.
Figure 1: Three users accessing a public cloud
Some of the best-known examples of public cloud systems are Amazon Web Services (AWS) containing the Elastic Compute Cloud (EC2) and the Simple Storage Service (S3) which form an IaaS cloud offering and the Google App Engine with provides a PaaS to its customers. The customer relationship management (CRM) solution Salesforce.com is the best-known example in the area of SaaS cloud offerings.