The on-demand, self-service, pay-by-use nature of cloud computing is essentially an extension of established trends in computing.   From an enterprise perspective, the on-demand nature of cloud computing helps to support the performance and capacity aspects of service-level objectives while the self-service nature of cloud computing allows organizations to create elastic environments that expand and contract based on the workload and target performance parameters. Furthermore, the pay-by-use nature of cloud computing may take the form of equipment leases that guarantee a minimum level of service from a cloud provider. 

Virtualization is a key feature of this model as virtualization enables organisations to easily and rapidly create copies of existing environments to support the development activities of testing, development and deployment. Often multiple virtual machines are involved. Minimal cost is involved as different systems and environments can coexist on the same servers as production environments and few resources are used.

 Likewise, new applications can be developed and deployed in new virtual machines on existing servers, opened up for use on the Internet, and scaled if the application is successful in the marketplace.  This lightweight deployment model has already led to a “Darwinistic” approach to business development as beta versions of software are made public and it is the market that effectively decides which applications will succeed or fail. The latter applications will be retired while the former will be scaled.

 Cloud computing extends this trend through automation.  Instead of negotiating with an IT organization for resources on which to deploy an application, a compute cloud is a self-service proposition where compute cycles can be purchased and a web interface or Application Programming Interface (API) is used to create virtual machines and establish network relationships between them. Therefore, instead of requiring a long-term contract for services with an IT organization or a service provider, clouds work on a pay-by-use, or pay-by-the-sip model where an application may   exist to run a job for a few minutes or hours, or    it may exist to provide services to customers on a long-term basis.

 Compute clouds are built as if applications are temporary, and billing is based on resource consumption, for example, CPU hours used, volumes of data moved or gigabytes of data stored. The ability to use and pay for only the resources used shifts the risk of how much infrastructure to purchase from the organization developing the application to the cloud provider.  It also shifts the responsibility for architectural decisions from application architects to developers. This shift can increase risk which must be managed by enterprises that have processes in place for a reason.   System, network, and storage architects need to factor in this risk to cloud computing designs.

So how does this model impact smart networks and smart objects? Given that may of the networks that may be ad-hoc or spontaneous in nature it follows that the processing of data and hence the consumption of resources will vary over time and depending on the network. The pay as you use model for computing is hence very beneficial for these networks as the number of devices and the demand on computing resources will vary. This may also apply even where the number of devices does not vary as data processing requirements will change over time or even over the course of a day for networks such as the smart grid.