Cloudware Basics - Guide to Cloud Computing for Business and Technology Managers - page 273

latest technologically upgraded and enhanced resources can be accom-
plished at relatively small cost (with minimal cost of replacing discontin-
ued resources) and offers the client access to advanced technologies that
are constantly being acquired by the cloud provider. Improved delivery of
services obtained by rapid cloud provisioning improves time to market and,
hence, market growth.
13.5.5 Scalability
Cloud computing provides the means, within limits, for a client to rap-
idly provision computational resources to meet increases or decreases in
demand. Cloud scalability provides for optimal resources so that computing
resources are provisioned per requirements seamlessly ensuring maximum
cost-benefit to the clients. Since the cloud provider operates on a multitenancy
utility model, the client organization has to pay only for the resources it is
using at any particular time.
SCALABILITY OF COMPLEX SYSTEMS LIKE THE
INTERNET
Complex systems exhibit different patterns of behavior than tra-
ditional systems and require new design principles based on a
deeper understanding of the physical properties of their components
and of the manner in which they interact with one another, and with the
environment. The behavior of any system cannot be explained without
precise knowledge of the interactions among its components, hence,
such an effort is undertaken whenever we need to make progress in
our understanding of the behavior of the physical world surrounding
us or a system that we engineer. Predicting all possible interactions
among the components of a system during the design process is an
even more daunting task.
The topology of a network used to model the interactions in complex
biological, social, economic, and computing systems is described by
means of graphs in which vertices represent the entities and edges rep-
resent their interactions. The number of edges incident upon a vertex
is called the degree of the vertex. Systems interconnected by scale-free
networks have the property that we can limit the number of interac-
tion paths among its components without limiting the number of com-
ponents of this network system. Because the degrees of nodes obey a
power-law distribution, while the vast majority of components have
one or very few connections, the number of highly connected compo-
nents is very small. This ensures and assures the scalability of scale-
free networks like the Internet.