Part2 - Cloud Computing Cost-Benefit Analysis Assessing Green IT Benefits

Motivations

Both cloud computing and green IT topics discussed in this dissertation address two of the most important epochal challenges and business opportunities of our time.

·        Cloud computing represents a paradigm shift that has the potential to disrupt the overall industry by displacing the traditional on-premises datacenter computing style toward a cloud-based computing style delivered as a service over the Internet, where company business processes should gain in flexibility and cost. This prospect is dizzying, but ultimately a very inspiring innovation and technology management business topic.

·        Understanding to what extent cloud computing can help mitigate the devastating effects of global warming through a reduction of the ICT industry carbon footprint, and how businesses can benefit from it, is yet another most inspiring business management topic.

Here is why:

Operating system and middleware software vendors like Microsoft and Sun Microsystems, perceive cloud computing as a disruptive innovation to their business as it is accelerating the software commoditization trend that started in the nineties with the open-source software phenomenon - such as the Linux operating system - that runs most of the servers delivering content on the Internet today. Likewise, makers of enterprise software, such as SAP and Oracle, may suffer similar market tensions with the advent of cloud computing. Up until now, they have made billions by selling very expensive software solutions, demanding hefty sums for their installation and then charging annual maintenance fees for upgrades and technical support and given business growth to service providers like Capgemini (the company I currently work for). But his lucrative business, says Michael Cusumano, professor at the Massachusetts Institute of Technology (MICT), in (Siegel 2008, p.10), has come under increasing pressure. The corporate world has become less and less willing to buy software for large sums of money, which explains why large independent software vendors (ISVs) have steadily increased their maintenance and other service fees over the years, and will continue to do so to maintain their revenues.

The biggest challenge for those firms will be to become providers of cloud-based online services themselves. So far, they have moved slowly, offering software-as-a-service (SaaS) solutions partly because their customers were not ready for a bigger move, but more importantly, because software firms are still fixated on their old business models, which call for upfront payment as opposed to a deferred pay-as-you-go revenue model (Siegel 2008, pp.10-11). This positioning became obvious through several discussions I recently had on cloud computing with engineering managers1 at Oracle. It appears that traditional ISVs, like Oracle, may not have the right resources, processes and values to compete effectively on the cloud- computing market battle-field.

Similarly, in a special report of The Economist on Corporate IT, Siegel Ludwig (2008) argues that, at the beginning, the biggest winners of the cloud computing market share battle are likely to be the hardware manufacturers. In the longer term, however, there will be relative winners and losers. “The hardware business could actually find itself in the losing group; its margins could get squeezed as this industry matures because there will be fewer customers with more bargaining power” (Siegel 2008, p.10), and that “in the long term, hardware manufacturers may be torn between supplying cloud providers or becoming providers themselves. Being both will not be easy because the firms may find themselves competing with their biggest customers” (Siegel 2008, p.10).

On the green IT front, a 2006 report by Sir Nicholas Stern, former head of the UK Government's Economic Services and Chief Economist of the World Bank, also known as the Stern Review, predicts that climate change will have a serious impact on economic growth without mitigation. The report suggests that failure to invest at least 1% to 2% of the UK gross domestic product (GDP) to mitigate the effects of climate change incurs the chance of a recession worth up to twenty percent of the global GDP (Bowen et al. n.d., p.VII) and (Van der Perre et al. 2009, p.14). Furthermore, the Intergovernmental Panel on Climate Change (IPCC) concluded that most of the observed global warming since the middle of the 20th century has been caused by increasing concentrations of greenhouse gases resulting from human activity such as fossil fuel burning and deforestation (Alley et al. n.d., p.1). Greenhouse gas emissions are forecast to grow by 130% between now and 2050, unless we learn to generate and use energy more efficiently. The International Energy Agency (IEA) calculates that to reduce CO2 emissions to half of today’s levels would require investing $45 trillion in the development and deployment of carbon emissions reduction technologies over this period, the equivalent of 1.1% of the average annual global GDP (Kanter 2008).

In “Green IT Going for Green”, VMWare and Intel (Speedfire et al. n.d.) throw a compilation of striking figures providing evidence that the ICT industry bears responsibility in the current global warming situation, and showing, for example, the dreadful impact of our seemingly benign attitude when consuming electronic goods on the Internet. For instance, downloading the electronic version of our daily newspaper uses the same amount of electricity as a laundry cycle in our clothes washer, according to the IZT research institute. Also, according to Gartner in (Speedfire et al. n.d.) the ICT industry is responsible for as much greenhouse gas as the world's airline industry, which contributes from 2% to 5% of the worldwide CO2 emissions3. This worrisome number draws from the fact that, in most developed and emerging nations, fossil fuel is used to produce electricity. Among these nations stand the US who, operating some of the largest datacenters in the world, produces 50% of its electricity out of burning coal. McKinsey & Co. in (GreenIT n.d.) forecasts that the ICT sector’s CO2 emissions will triple during the period from 2002 to 2020, and that in office buildings ICT typically accounts for more than 20% of the energy used, and in some offices up to 70%.

While there is little academic research to support the claim that cloud computing has a positive environmental impact, there is reason to believe that it could help the ICT industry reduce its carbon footprint. The basic reasons for this are:

·        Cloud computing provides a much higher utilization rate of ICT resources than conventional datacenters.

·        Cloud providers' best interests are to optimize their energy consumption because this latter directly affects their profit margins.

·        The industrialization and maturity of the cloud computing market will draw further consolidations and improve ICT efficiency, leading to a reduction in energy requirements.

Approach

CBA will produce reasonably accurate results only as long as it is used in the right way. As such, this dissertation employs a multi-stepped approach to CBA proposed by (Baker et al. n.d.) and by (Hanley & Spash 1995). (Baker et al. n.d.) proposes the case of a CBA methodology for IT related projects. Hanley and Spash in (1995) provides an introduction to CBA—and the closely related technique of cost-effectiveness analysis (CEA)—for people interested in its application to environmental management. In this book, they have tried to represent most sides of the arguments concerning contentious issues around the correct approach to appraising projects that may impact positively or negatively the environment. Both works propose a CBA structure that is roughly outlined in six essential steps.

Determine the objective and scope
This step establishes an understanding of the business problem the ICT project is trying to solve. It is important to keep in mind that businesses do not implement technologies for their own sake. In other words, businesses will implement a new solution to solve a particular problem, improve a certain process, become more efficient, and so forth. This step will document the business drivers that make the case for the ICT project under consideration, and list the deliverables that support the business drivers. In this step, I will also provide the baseline cost of business, meaning what the company spends today in the domain under which the ICT project will operate. Recalling the business drivers mentioned above, the baseline cost of business should highlight how much the company currently spends on maintaining the information system that support its business processes. Determining the baseline is important to effectively calculate the magnitude of the benefits of the new ICT project.

Determine which costs and benefits are economically relevant 
This step involves a review of all the costs and benefits that are economically relevant to the project. From an environmental point of view, the economically relevant impacts—what to count—that can be accounted for in a CBA are necessary so long as they are measurable within the scope of this study, even though the theory says that “the environmental impacts count so long as they either: 1) cause at least one person in the relevant population to become more or less happy; 2) change the level or quality of output of some positivity valued commodity” (Hanley & Spash 1995, pp.8-20). For example, the positive impact of Google's Dalles datacenter that uses the cold waters of the Columbia river for its cooling system may save on carbon-dioxide (CO2) emissions, but may adversely deteriorate the landscape. In theory, this negative effect is relevant to an environmental CBA as at least one person will most likely dislike the landscape change because the absence of a market for landscape quality is irrelevant here. In fact, many environmental effects fail to be recorded by market price movements. Nevertheless, unpriced impacts—referred to as externalities—are the most important feature of environmental CBA according to (Hanley & Spash 1995, p.10). The central message here is that, while most environmental impacts are likely to be relevant in the context of a full- fledge environmental CBA, they will not necessarily be carried out in that study for obvious lack of measurability reasons. On the other hand, they may induce additional project choice incentives in the conclusion.

Determine the project costs (monetary valuation of the negative effects) 
Once I have identified which costs are relevant to the project, I will need to calculate their ex ante values in monetary terms. There are two important things to keep in mind according to (Baker et al. n.d., p.2):

·        Ensure that the analysis includes all the cost categories.

·        Express how much it will cost to maintain the system after implementation. This is known as the full life-cycle cost.

Determine the project benefits (monetary valuation of the positive effects) 
Similarly to determining the project costs, I will need to provide a list of the benefits that the project will gain from the solution. For each of these benefits, I will need to provide a quantification method to be able to calculate their ex-ante values in monetary terms.

Discounting of cost and benefit flows
 The project's net cash stream will contain a dollar value for each year starting with the first year of the project and ending after 3 years. The cash-stream schedule indicates the timing of the costs and benefits which result in a positive or negative net cash flow. Once all relevant cost and benefit flows have been so expressed, it is necessary to convert them into present value (PV) terms. This step stems from the need to appraise cost and benefit flows into their present value (PV) terms to take into account the time value of money. The project cash-stream is the cornerstone of the financial analysis. From it, I will be able to calculate the Return on Investment (ROI), the payback and Net Present Value (NPV).

Applying the Net Present Value Test
The main purpose of the CBA is to help select projects and policies that will be the most efficient in terms of their use of resources. The criterion applied is the Net Present Value (NPV) test. This test simply asks whether the sum of the discounted gain exceeds the sum of the discounted losses. If so, the project can be said to present an efficient shift in resource allocation, given the data used in the CBA.