The key concepts of cloud computing

 we are bombarded daily with ads about products, white papers, publications, seminars, workshops, market projections and more focused on cloud computing. But if you analyze these messages you reach the conclusion there are many ambiguities and only a few emerging concrete terms.

For a start, cloud computing is not a technology but a model of provision and marketing IT services that meet certain characteristics. Cloud is all about computer services, not products:

* The infrastructure is shared. Multiple clients share a common technology platform and even a single application instance.

* The services are accessed on demand in units that vary by service. Units can be, for example, user, capacity, transaction or any combination thereof.

* Services are scalable. From the user's point of view, services are flexible; there are no limits to growth.

* The pricing model is by consumption. Instead of paying the fixed costs of a service sized to handle peak usage, you pay a variable cost per unit consumption (users, transactions, capacity, etc.) that is measured in time periods that can vary, such as  hour or month.

* Services can be accessed from anywhere in the world by multiple devices. The cloud model leads to basically two different kinds of clouds: private and public. The public clouds are those that offer IT services to any customer over the Internet. Private clouds offer IT services to a predefined group of customers, with access through Internet or private networks. You might have also heard about internal and external clouds. The former are a subgroup of the private clouds, and provide services within the same company or corporate group. The latter may be public or private and provide services to other companies.

FAQ: Cloud computing, demystified

IT services provided through the cloud are grouped into three categories: infrastructure-as-a-service (IaaS), platform-as-a-service (PaaS) and software-as-a-service (SaaS).

IaaS provides the processing environment (servers, storage, load balancers, firewalls). These services can be implemented through different technologies, virtualization being the most common one, but there are implementations that use grid technologies or clusters.

PaaS provides an environment for developing and running applications. Authentication, authorization, session management and metadata are also part of this service.

SaaS is the most advanced and complex cloud model. The software services provide functionalities that solve user problems, whether it's an individual or an employee of a company. Some examples of solutions that are now offered under the SaaS model include: business intelligence, Web conference, e-mail, office automation suites and sales force automation.

The benefits of this model are clear and very attractive: to access a comprehensive service, eliminate investments, defer some costs and eliminate others, increase agility of IT areas, increase user mobility and improve the availability of services.

However, given the novelty of the model, there are some aspects that have not yet been resolved, and as in all work environments, there are risks that must be taken into account when assessing how, when and for what to use this new tool that is available to IT areas of companies.

Current implementations of the cloud computing model do not offer adequate service-level agreements or their control tools yet. They do not provide safety audit process or regulations for storage and backup of customer data that are managed and hosted in the cloud. Nor do they provide integration interfaces that are clear and sustainable over time, between the services taken from the cloud and those of the company itself.

Like any tool, cloud computing is not perfect, nor is it applicable in all settings and it can't be implemented overnight. It requires an evaluation process of benefits and risks, a phased plan for implementation, and defining a process of continuous improvement for the production stage.

Given the weaknesses that have been mentioned, cloud services are an excellent choice, in this first period, for businesses to implement private clouds using IaaS, for testing environments or application development in the PaaS mode and for self-contained applications, such as Web conference in the SaaS category.

While the cloud computing model presents new options for users, it also demands significant changes from product providers and IT services.

Companies providing hardware, software and services will have to face two challenges simultaneously. The first is to develop or adapt their products, processes and tools to participate in this new IT marketing and provision model. The second, and perhaps the hardest, is to change their sales and support structures, and in some cases, their target markets as well, in order to meet the dynamics of the new environment.

For companies that sell base hardware and software, the main change will be at the level of their target market: in a cloud computing model the technology consumers are no longer the financial companies, industries or state organisms -- which become users -- but the companies providing IT services.

For companies that sell applications, the impact is going to take place in the sales and support structure, for they need to move from selling licenses to selling and supporting services.

Finally, companies providing IT services will have to integrate communications to their services at a 100%, adapt their portfolio to the cloud category in which they decide to participate (IaaS, PaaS or SaaS), adapt their sales, delivery and billing processes to the on-demand model, and develop the necessary integration mechanisms between the customer's IT area and the cloud to make the development of the model possible.

Given the scale and complexity of the changes lying ahead, it isn't a surprise to have several false starts. The important thing is to devise a plan, understanding at each step which the challenges to be faced are and evaluating the risks to take, and to begin in order to avoid delaying or discarding any alternative to make the business more profitable and efficient.

Read more about data center in Network World's Data Center section.