A View of the Cloud from the Ground :)

While there exists an overwhelming plethora of business- and technical-related articles, analyses and reports about cloud computing, there is surprisingly little academic studies devoted to this phenomenon. As stated earlier, confusion among ICT practitioners still abounds when cloud computing is discussed. For example, many people believe that there will be thousands of public clouds, while Gartner, for instance, recognizes that there is only one cloud that they call “The World Wide Computer”. Certainly, there are many cloud platforms, such as Amazon's EC2, and many cloud services from companies ranging from Google to Zoho; all of them will inject their services into the one public cloud. Some believe that SaaS and cloud computing are exactly the same, while others do not. In its controversial report “Clearing the air on cloud computing”, McKinsey & Co. argues that the lack of sharing a common understanding and rigorous definition of cloud computing prevents managers from being analytical in their decision-making and from making more informed investment decisions. A clearer definition of cloud computing would allow service providers to build more meaningful products, marketing and sales strategies that would translate into real value for their customers.

Cloud computing is one of those terms whose definition and interpretation vary the most between groups of people who have different perspectives and varying degrees of understanding and exposure. The variety of technologies and the hype surrounding the concept, as stated earlier, reinforces the overall confusion about the paradigm and its capacities, turning the cloud into an excessively used general term that includes almost any solution that allows the outsourcing of all kinds of ICT services. Another source of confusion highlighted by Vaquero et al. in (2008) lies in the relation of cloud computing o grid computing. “The distinctions are not clear, maybe because clouds and grids share similar visions: reduce computing costs and increase flexibility and reliability by using third-party operated hardware.” (Vaquero et al. 2008, p.51). However, cloud, grid and utility computing differ in subtle ways. Cloud computing includes attributes previously associated with utility and grid computing models. Grid computing refers to the ability to harness large collections of independent and heterogeneous computing resources to perform large tasks, and utility is a metered consumption of ICT services. Cloud computing is taking these attributes together, creating a more exiting ICT service delivery value proposition, says Kristof Kloeckner, the cloud computing software chief at IBM in (Brodkin 2009).

Smith et al. in (2009) also believe that it is important to dig beyond the general cloud computing concept to separate the hype from the actual ideas and technology benefits. To understand them it is necessary to tear apart the hype surrounding cloud computing by focusing on more granular topics which are part of the cloud phenomenon.

The following Section explores the concept in greater detail as confusion and conflicting definitions may lead to the risk of overlooking what business benefits can be obtained from this paradigm shift. This exploration includes an analysis of the business and technological advances underpinning the concept in an attempt to delineate the scope of our research and focus primarily on the utility dimension of cloud computing and on how it partakes with green IT to achieve the expected business benefits of our project.

Defining the Concept

This Section presents some notable attempts at rigorously defining cloud computing and capturing its key characteristics. It is important to keep in mind that the concept is in constant evolution, so any definition is worth only for how cloud computing is perceived today. Its key characteristics, deployment models, risks and benefits will be refined over time as the technology and business models mature, and knowing also that the ICT industry represents a large ecosystem of business models, vendors and market niches.

Nonetheless, any definition should attempt to encompass all of the various cloud approaches.

Vaquero et al. in report that there are over 20 different definitions of cloud computing in literature today. They argue that these definitions seem to merely focus on certain aspects of the technology and that none of them put forward a completely integrative definition. Thus, they have tried to reach a definition that encompasses all of them by extracting relevant cloud features and combining them to form both an integrative and basic cloud definition containing the essential characteristics of the technology. Among these definitions, Vaquero  et al. claim that theone proposed by Jeremy Geelan in “Twenty one experts define cloud computing” (Geelan 2009) is particularly interesting because it gathers definitions proposed by many experts, even though his definition lacks a more global analysis of those proposals to reach a more comprehensive definition. A minimalist definition of cloud computing by Vaquero et al. stands as a“Pay-per-use utility model and virtualization”

A minimalist definition such as the above, looking for a minimum common denominator, brings little value. A more useful, yet unified and comprehensive, definition of cloud computing by Vaquero et al. is substantially more thorough.

Armbrust et al. (2009) from the RADS Laboratory of UC Berkeley take a slightly different approach to defining cloud computing in “Above the clouds: A Berkeley view of cloud computing”, a well-regarded academic paper on the subject. They argue that the term cloud computing refers to both:

·        Applications delivered as services over the Internet.

·        The hardware and systems software datacenters that carry these services' runtime.

These two bullet points can be contracted as “self-service ICT delivered through automation”. They also claim that a cloud is different from a conventional datacenter in several ways. From a hardware point of view, at least two aspects are new with the cloud.

·        Cloud platforms bring the illusion of infinite computing resources available on demand, thereby eliminating the need to plan capacity growth ahead of time.

·        Cloud platforms eliminate upfront investments as companies of all sizes can provision ICT resources on demand, with no upfront expenses or long-term commitments, and pay only for the resources they use. The ability to pay for resources on a short-term basis as needed saves energy by freeing machines and storage devices when they are no longer useful.

When a cloud is made available to the public in a pay-as-you-go manner, it is referred to as a public cloud, and the service being offered as utility computing.  Classic examples of public clouds include Amazon Web Services (AWS), Google AppEngine, and Microsoft Azure.

McKinsey & Co. argues that cloud services known as SaaS may be confused with true clouds when they comply only with one or two of the above key requirements, but not all three. Typically, McKinsey & Co. would not qualify Gmail and Salesforce.com services as cloud-based services because they do not comply with the infrastructure cost as variable OPEX requirements.

Finally, I will conclude with the definition of cloud computing by the National Institute of Standards and Technology (NIST), which is the American equivalent of the Association  Française de Normalisation (AFNOR), the French national organization for standardization.

In reality, it is unclear whether there is such a thing as a canonical, standard and commonly agreed definition of cloud computing because it is a complex and continuously evolving concept. To fully understand it, we need to describe it as an overall paradigm shift within which emerge some essential and unique characteristics, as well as the service and deployment models that underpin the multiple ways cloud computing is being brought to market. 

As Capgemini's company bus don't wait for me if I am late :) , so later on but soon I will describes the three most common service models known to cloud computing today, and will provide a description of its commonly agreed fundamental characteristics and deployment models.