The Cloud Adoption Toolkit: Addressing the Challenges of Cloud Adoption in Enterprise

The Cloud Adoption Toolkit: Addressing the Challenges of Cloud Adoption in th e Enterprise Ali Khajeh-Hosseini, David Gree nwood, James W. Smith, Ian Somm erville Cloud Computing Co-laboratory, School of Computer Science, University of St Andrews, UK {akh, dsg22, jws7, ifs}@cs.st-andrews.ac.uk Abstract. Cloud computing promises a radical shift in the provisioning of computing resource within the enterprise. This paper: i) describes the challenges that decision makers face when attempting to determine the feasibility of the adoption of cloud computing in their organisations; ii) illustrates a lack of existing work to address the feasibility challenges of cl oud adoption in the enterprise; iii) introduces the Cloud Adoption Tool kit that provides a framework to support decision makers in identifying their conc erns, and matching these concerns to appropriate tools/techniques that can be used to address them. The paper adopts a position paper methodology such that case study evidence is provide d, where available, to support claims. We conc lude that the Cloud Adoption Tool kit, whilst still under development, shows signs that it is a useful tool for decision makers as it helps address the feasibility challenges of cloud adoption in the enterprise. Keywords: Cloud computing, cloud adoption, decision support 1 Introduction Cloud computing is the latest effort in delivering computing resources as a service. It represents a shift away from computing as a product that is owned, to computing as a service that is delivered to con sumers over the internet from large -scale data centres – or ‗clouds‘. Cloud computing is currently being exploited by technology start -ups due to its marketed properties of scalability, reliability and cost -effectiveness. Enterprises are also beginning to show an interest in cloud computing due to these promised benefits, however at present much ambiguity and uncertainty exists regarding the actual realisability of these promised benefits in the enterprise. Whilst there is much hype surrounding cloud computing, particularly around it s cost savings which are based on simplistic assumptions, we believe the technology is still likely to have a profound effect on the ways software will be procured, developed and deployed, similar to the e ffect of moving from mainframes to PCs. This paper‘s original contribution is to propose the Cloud Adoption Toolkit, which provides a collection of tools that can be used to support decision making during the adoption of cloud computing in the enterprise. Our toolkit is based on a framework to organise thin king about decision makers ‘ concerns an d mat ch these to tools that 2 Ali Khajeh-Hosseini, David Greenw ood, James W. S mith, Ian Sommerville address these con cerns, where each tool enables decision makers to focus on and model different attributes of their organisation s or IT systems. These models ca n then be used to reason about and investigate cloud adoption decisions. For example, by modelling a sy stem‘s har dware infrastructure and applications, it becomes possible to estimate the costs of running that system in a cl oud, and hence decide whether de ploying that system in the cloud would be cost effective. Furthermore, by identifying the impacts of a proposed system to people's work activities, the proposed system‘s practical and socio -political feasibility can be determined. For example, a system may be cost effective yet socio -politically infeasible if it potentially decreases job satisfaction and underm ines existing power bases or organisational values [1] . This paper is structured as follows: Section 2 introduces cloud computing and discusses the importance of modelling in systems engineering; Section 3 describes the challenges of cloud adoption in the enterprise and illustrate s the challenges by looking at a case study; Section 4 describes the conceptual framework behind the Cloud Adoption Toolkit and provides details of its individual tools; and Section 5 concludes that the toolkit is prom ising and presents our future wo rk. 2 Background 2.1 Cloud Computing There are many definitions of cloud computing [e.g. 2, 3, 4]. The US National Institute of Standards and Technology (NIST) has published a working definition that has captured the commonly agreed aspects of clou d computing; it defines cloud computing as ― a model for enabling convenient, on-demand network access to a shared pool of co nfigurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction ‖ [5]. The NIST definition describes cloud computing as being c omposed of:  Five characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service.  Four deployment models: private clouds, community clouds, public clouds, and hybrid clouds.  Three service models: So ftware as a Service (SaaS), Platform as a Serv ice (PaaS), and Infrastructure as a Servic e (IaaS). Figure 1 provides an overview of the common deployment and service models in cloud computing, where the three service models could be deployed on top of any of the four deployment m odels. Fig. 1. Cloud computing de ployment and service m odels The Cloud Adoption Toolkit: Add ressing the Challeng es of Cloud Adoption in the Enterprise 3 2.2 Modelling Making decisions regarding implementations of enterprise IT systems is complicated due to the intricate interrelationships between peop le and technology that decision makers must con tend with. Successful systems are a consequence of a stabilised mesh of sta keholder interests, values, know-how and technological characteristics [6, 7] . Modelling has an important role to play in the engineering of successful enterprise systems because it enable s decision makers to make sense of a nd represent these intricate interr elationships [8, 9]. Successful models address two challenges; the representational challenge and the informational challenge. The representation challenge is concerned with how to best represent the information included in the model to make it easy to understand. The informational challenge comprises what to leave-in and what to leave -out of a model. The informational challenge can be extremely challenging in the field of systems engineering as some systems such as so cial systems are hard to define and delimit. In order to address the inf ormational c hallenge of complex systems, frameworks or paradigms can be employed to reduce complexity by introducing simplifying assumptions. Modelling IT infrastructure is particularly challenging as enterprise IT environments have many variables and these variables have multiple interactions such that knowledge of their causal intricacies is especially important for selecting appropriate modelling primitives. One important way of managing this modelling challenge is by using frameworks, based upon rigorous studies of the dynamics of the domain and experiential knowledge wh ere rigorous studies are unavailable, to select appropriate primitives [6 , 7]. 3 Challenges of Cloud Adoption Cloud computing is not simply about a technological improvement of data centres but a fundamental change in how IT is provisioned and used [11]. Enterprises need to consider the benefits, risks and the effects of cloud computing on their organizations and usage-practices in order to make decisions about its adoption and use [12]. In the enterprise, the ―adoption of cloud computing is as much dependent on the maturity of organisational and cultural (including legislative) processes as the technology, per se" [13]. The a doption of cloud computing is not going to happen overnight – some predict that it could take between 10 to 15 years before the typical enterprise makes this shift [14]. Therefore we are currently at the start of a transition period during which many decisions need to be made with respect to cloud adoption in the enterprise. Cloud adoption decisions are challenging because of a range of practical and socio- political reasons. It is unlikely that all organisations will completely outsource their back-end computing requirements to a cloud service provider. Rather, they will establish heterogeneous computing environments based on dedicated servers, organisational clouds and possibly more than one public cloud provider. How their 4 Ali Khajeh-Hosseini, David Greenw ood, James W. S mith, Ian Sommerville application portfolio is distributed across this environment depends not just on technical issues but also on socio-technical factors (e.g., concerns about costs, confidentiality, and control), the impact on work practices and constraints derived from existing business models. Therefore, the challenges that a cloud adoption toolkit must address are: i) to provide accurate information on costs of cloud adoption; ii ) to support risk management; and iii ) to ensure that decision makers can make informed trade-offs between the bene fits and risks. It is important that decision makers take into account these considerations because the use of cloud computing has significant implications for an enterprise as a whole [1] . For example, we recently performed a feasibility analysis of a proposed cloud- based IT system at an SME in the oil and gas industry [1] . We found that despite the promised financial benefits, opportunities to remove tedious work from IT staff and the potential to enter new marketplaces, almost all of the stakeholder groups were neutral or reluctant to support a move to the cloud due to concerns regarding its impact on their work, increased risk of dependence upon third par ties and its implications for custom er service and support. A recent review of the academic research done in cloud computing revealed that there are currently no mat ure techniques or toolkits available to support decision making during the adoption of cloud computing in the enterprise [12, 15] . In industry, [16] and [17] provide examples of typical of ferings from IT consultancies that attempt to fill this gap. Such approaches have two problems: they are based on closed proprietary tools that are not widely available; and they are often accompanied by expensive consultancy periods. In contrast, we argue that given the Cloud Adoption Toolkit, enterprises can assess the feasibility of using cloud computing in their organisations themselves. However, the t oolkit can also be used by decision makers to verify the claim s made by IT consultancies and cloud servi ce providers. 4 The Cloud Adoption Toolkit The Cloud Adoption Toolkit comprises a conceptual framework for organizing decision makers ‘ concerns and matching the se to tools that address the m. Decision makers can use any tools/techniques that they wish to, however, we provide five tools/techniques that we believe to be extremely useful: Technology Suitability Analysis; Cost Modelling; Energy Consumption Analysis; Stakeholder Impact Analysis; and Responsibility Modelling. 4.1 Conceptual Framework The purpose of the conceptual fr amework is to organise decision makers ‘ thinking about the concerns that they and other stakeholders have, and the tools that can be used to explore the se concerns. It is important that decision makers view the proposed cloud adoption project from multiple stakeholders‘ perspectives in order to learn from a div erse range of stakeholder concerns and receive a broad range of feedback from the organisational environment . Figure 2 provides an overview of the Cloud Adoption The Cloud Adoption Toolkit: Address ing the Challeng es of Cloud Adoption in the Enterprise 5 Toolkit and how it can be used. Decision makers would start with a Technology Suitability Analysis, and if the cloud is found to be suitable for their system, they would pr oceed by investigating eithe r the costs of running the syst em on public clouds, or the energy consumption (and hence energy costs) of running the system on private clouds. At the same time, Stakeholder Impact Analysis can be performed to assess the impacts of using cloud com puting on the work of stakeholders in the enterprise. If these analyses show that running the system on the cloud is a viable option, then Responsibility Modelling can be performed to identify and analyse the risks associated with the operation of the system on the cloud, where diff erent cloud providers could be responsibl e for different aspects of the syst e m. Fig. 1. Cloud adoption conceptual framework 4.2 Technology Suitability An alysis The purpose of Technology Suitability Analysis is to support decision makers in determining whether cloud com puting exhibits the appropriate technological characteristics to support their proposed system. Understanding the characteristics of cloud computing is extremely important as it has the potential to exhibit radically different properties to those of traditional enterprise dat a centres. This is mainly due to the cloud ‘ s highly scalable nature, physi cal resource sharing between virtual machines, potential issues to do with communication over the internet and insufficient guarantees regarding the up-time and reliability of processing and data storage services. For example, typical IaaS offerings make no reassuring guarantees about server uptime or network performance wh ich has important implications for the viability of certain classes of software architectures and business c ritical systems. The Technology Suitability Analysis comprises a simple checklist of questions to provide a rapid assessment o f the potent suitability of a particular cloud service for a specific enterprise IT system. The checklist is still under development bu t the current 6 Ali Khajeh-Hosseini, David Gre enwood, James W. S mith, Ian Sommerville version, shown in Table 1, analyses eight characteristics and quickly prov ides an indicat ion of the cloud‘s suitabilit y for a proposed IT system . Table 1. Technological Suitability Analysis Desired Technology Characteristic Questions 1. Elasticity - Does your software architecture support scaling out? - If not, will scaling up to a bigger server suffice? 2. Communications - Is the bandwidth within the cloud and between the cloud and other systems sufficient for your application? - Is latency of data transfer to the cloud acceptable? 3. Processing - Is the CPU power of instances appropriate for your application at the expected operating load? - Do server instances have enough memory for your application? 4. Access to hardware / bespoke hardware - Does y our cloud provider provide the required access to hardware components or bespoke hardware? 5. Availability / dependability - Does your cloud provider provide an appropriate SLA? - Are you able to create the appropriate availability by mixing geographical locations or service providers? 6. Security requirements - Does your cloud service provider meet y our securi ty requirements? (e.g. do they support multi-factor authentication or encrypted data transfer) 7. Data confidentiality and privacy - Does your cloud provider provide sufficient data confidentiality and privacy guarantees ? 8. Regulatory requirements - Does y our cloud provide r comply with the required regulatory requirements of your organisation? 4.3 Cost Modelling The purpose of the Cost Modelling tool is to support cloud adoption decisions in two different ways: 1. To support decision makers in obtaining accurate cost estimates of running IT systems on the cloud. The tool hel ps decision makers investigate the costs of migrating an existing IT system or deploying a new IT system on the cloud, the costs of migrating an IT system from on e cloud to another, or even future costs based upon predictions of fut ure workload. 2. To support system architects in evaluating the design of a proposed IT system with respect to its operational cost s, with the aim of m inimizing the costs. Simplistically, IT system pr ocurement is based on obtaining estimates for a proposed system, then getting those estimates signed -off by management or the client to allow the procurement to proceed. Capital and operational budgets are often kept separate in many organizations, and procurement costs have to be known in advance before approval can be gained. Currently, the estimation process is based on The Cloud Adoption Toolkit: Address ing the Challeng es of Cloud Adoption in the Enterprise 7 predicting the maximum quan tity of resources (i.e. processing power, memory, storage etc.) that a system might need and provisioning at that level. However, most of the acquired resources remain unused during normal operation as the estimates were based on peak load. In fact recent figures show server utilization in traditional data centres ranging from 5% to 20% [2]. The utility billing model of cloud computing has a certain degree of uncertainty that goes against current procurement policies. The uncertainty relates to: i) the actual resources consumed by a system, which are determined by its load; ii) the deployment option used by a system, which can affect its costs as resources like bandwidth are more expensive between clouds compared to bandwidth within clouds; iii) the cloud service provider ‘s pricing model, which c an change at any time. The utility billing model is also a shift away from capital to operational budgeting, and many enterprises are less savvy about operational budgeting for IT than they are fo r capital budgeting. Cost Modelling extends the capab ilities of UML deployment diagrams [8], which enable a system‘s deployment to be modelled. In its essence, a UML deployment diagram enables users to model the deployment of software artefacts onto hardware nodes. The Cost Modelling tool enables us ers to model a system‘s software applications and how they could be deployed on cloud, traditional or hybrid infrastructures. The model is then processed to give users an accurate estimate of the operational costs of their system. The models can take in to account future resource demands therefore enabling for situations wh ere traditional infrastructure may not initially be cost-effective, yet will become cost effective with future workload increases . The concepts behind the Cost Mod elling tool were used in a recent case study that compared the infr astructure costs of deploying a system on Amazon‘s Elastic Compute Cloud (EC2) versus a traditional data centre. The tool showed that the system infrastructure in the case study would have cost around 37% less over 5 years on EC2 compared with a tradi tional data centre [1]. 4.4 Energy Consumption Analysis The purpose of Energy Consum ption Analysis is to support decision making regarding the optimum energy usage of IT resources when a system is deplo yed on a private cloud . This tool will help to inform decision making and aid the design of cloud-based architectures by enabling the assessment of potential trade-offs between energy efficiency and performance in order to build system s to particular requirements. This tool is currently under development and investigations into this area are ongoing . However, the idea of reducing a system‘s ener gy consumption is promising as a traditional blade server consumes 50% of its peak energy when it is switched on and doing only 10% of its capable work. This disproportionate amount of consumption is caused by the server‘s baseline consumption [ 18 ]. If a server is not being fully utilized then the baseline consumption will dominate the ov erall energy consumption of a system. Cloud computing provides an alternative to this situation as Infrastructure as a Service providers such as Amazon deploy multiple virtual servers on the same ph ysical server . For example, on the StACC Private Cloud (www.cs.st- andrews.ac.uk/stacc) w e use a mapping of 8 virtual servers to one physical machine. 8 Ali Khajeh-Hosseini, David Greenw ood, James W. S mith, Ian Sommerville Virtualisation techn iques enable systems to achieve far higher levels of utilisation, and therefore can save energy by reducing the number of physical servers that need to be operating. 4.5 Stakeholder Impact Analysi s The purpose of Stakeholder Impact Analysi s is to support decision makers in determining the socio-political viability, or benefits and risks, of a proposed IT system. This is i mportant as cloud adoption projects are not merely technolog ical upgrades but involve the reconfiguration of working practices and technologies to take full advantage of the benefits offered by the technology [11]. The socio-political benefits and risks associated with a proposed IT system are determined by identif ying the i mpact of changes to stakeholders ‘ work activities in terms of their pr acticalities (time, resources, and capabilities), so cial factors (interests, values, status, and satisfaction) and political factors (their perception of the fairness of decision making procedures and the distribution of ben efits, drawbacks and risks). This information en ables decision makers to make a judgement about the risk that specific stakeholders will hold unsupportive attitudes towards the proposed system and therefore indicates the overall socio-political feasibility of the system. Stakeholder Impact Analysis has been successfully used to support decision making regarding the feasibility of migrating a client -server application (a data acquisition and quality monitoring system) to Amazon EC2 [1] . The analysis revealed that the proposed cloud migration would have many implications for the organization including non-technical areas such as the finance and marketing departments. Overall a positive net benefit was perceived from the perspectives of the business development functions of the enterprise and the more junior levels of the IT support functions. A zero net benefit was perceived by the project management and support management functions of the enterprise and a neg ative net benefit was perceived by the technical manager and the support engineer functions of the ente rprise. The analysis i dentified numerous potential benefits and risks associated with the migration. Most notably, opportunities for improved cash flow management, opportunities to offer new products/services, and removal of tedious work were identified as benefits. In contrast the following notable risks were also elicited: the deterioration of customer care and service quality; increased dependency on external 3 rd parties; and departmental do wn -sizing. Ov erall the decision makers decided that for this parti cular situation the benefits/opportunities did not outweigh the drawbacks/risks and there fore decided a migration was not viable in the short term . 4.6 Responsibility Modelling The purpose of responsibility modelling is to support decision makers in determining the operational viability of a proposed IT system. R esponsibilit y modelling also helps decision makers in identifying and analysing risks associated with the operation of complex IT systems [19]. Responsibili ty modelling is particularly important for systems deployed on the cloud as the responsibilities for The Cloud Adoption Toolkit: Address ing the Challeng es of Cloud Adoption in the Ente rprise 9 constructing, operating, maintaining, and managing the system can be divided across multiple organisations, departments and cloud service prov iders, and therefore identifying and managing the risks associated with the discharge of responsibilities is important to the operational vi ability of the IT system . The viability of the system is determined by: i) identifying the set of responsibilities that must be discharged for the system to op erate according to a set of non-functional requirements; ii) who is responsible for what; iii) whether the configuration of responsibilities is likely to meet non-functional requirements of the system; and iv) determining the practical, social and political viability of the discharge of responsibilities so that the system exhibits appropriate non -functional characteristics e.g. up-time, responsiveness, resilience, maintainability and recoverability. 5 Conclusion Cloud computing is currently being exploited by technology start-ups due to its marketed properties of scalability, reliability and cost effectiveness. Enterprises are also beginning to show an interest in cloud computing due to these promised benefits, however at present much ambiguity and uncertainty exists regarding the actual realisablility of these prom ised benefits in the enterprise. This paper demonstrated that current feasibility approaches fall short in terms o f enabling enterprise decision makers to determine the viability of using cloud computing, and that our Cloud Adoption Toolkit offers a promising starting point. Our conclusions are limited due to position paper methodology that this paper uses and its reliance upon lim ited case study data. The Cloud Adoption Toolkit is designed such that it can incorporate new tools to account for emerging factors with respect to cloud computing, and in the f uture we aim to accommodate other industrial concerns such as how clou d computing can reduce an organisation‘s carbon emissions. W e are currently in the process of planning a number of industrial case studies to evaluate and further develop the Cloud Adoption Toolkit. Acknowledgements We thank the Scottish Informatics and Computer Science Alliance (SICSA) and the EPSRC for fu nding the authors. We also thank our colleagues at the UK‘s Large - Scale Complex IT System s Initiative (www.lscits.org) for their comm ents. References [1] A. Khajeh-Hosseini, D. Greenwood, and I. Sommerville, "Cloud Migration: A Case Study of Migrating an Enterprise IT Syst em to IaaS," To appear in the 3 rd International Conference on C loud Computing (IEEE CLOUD 20 10), (2010) 10 Ali Khajeh-Hosseini, David Greenwo od, James W. Smith, Ian So mmerville [2] M. Ar mbrust, A. Fox, R. Griffith, A. Joseph, R. Katz, A. Konwinski, G. Lee , D. Patterson, A. Rabkin, I. Stoica, and M. Zaharia, "Above the Clouds: A Berkeley View of Cloud Computing.," Technical Report, University of California at Ber keley (2009) [3] L. Vaquero, L. Merino, and J. Caceres, "A break in the clouds: towards a cloud definition," SIGCOMM Comp. Comm unications Review, vol. 39, p p. 50 — 55 (2009) [4] L. Youseff, M. Butrico, and D. Da Silva, "Toward a Unified Ontology of Cloud Computing," Grid Com puting Environments Workshop (GCE ' 08), pp. 1 — 10 (2008) [5] P. Mell and T. Grance, "The NIST Definition of Cloud Computing," National Institute of Standards a nd Technology (2009) [6] P. Checkland and S. Holwell, Information, Systems and Information Systems, Chichester: John Wiley and Sons Ltd ( 1997 ) [7] R. Kling, G. McKim, and A. Kin g, "A bit more to it: Scho larly communication forums as socio-technical interaction networks," Journal of the American Society for Information Science and Tech nology, vol. 54, pp. 47 — 67 (2003) [8] G. Booch, J. Rumbaugh, and I. Jacobson, Unified Modeling Language User Gui de, Addison-Wesley Professional (2005) [9] P. Beynon-Davies, Information Systems: An Introduction to Informatics in Organisations, Basingstoke: P algrave Macmillan (2002) [10] D. R. Anderson, D.J. Sweeney, and T.A. Williams, Quantitative Methods for Business, South-Western College P ub (2005) [11] M. Creeger, "CTO roundtable: cloud com puting," Comm. of the ACM, vol. 52, pp. 50 -- 56 (2009) [12] A. Khajeh-Hosseini, I. Sommerville, and I. Sriram, "Research Challenges for Enterprise Cloud Com puting," Unpublished, http://arxiv.org/abs/ 1001.3257 (2010) [13] W . Fellows, "Partly C loudy, Blue-Sky Thinking About Cloud Computing," Whitepaper. 451 Group ( 2008 ) [14] T. Sullivan, "The ways cloud computing will disrupt IT," http://www. cio.com.au/article/296892/nic k_carr_ways_cloud_computing_wil l_disrupt_it (2009) [15] I. Sriram and A. Khajeh -Hosseini, "Research Agenda in Cloud Technologies," Unpublished, http://a rxiv.org/abs/1001.3259 (2010) [16] Accenture, "Accenture C loud Computing Accelerator," http://www.accenture.com /Global/Services/Accenture_Tec hnology_Labs/R_and_I/Cl CloudComputingAccelerat.ht m (2009) [17] Computer Sciences Corporatio n, "Doing Business in The Cloud," http://www.csc.com/cloud/ds/ 35354-csc_cloud_adoption_assessm ent (2009) [18] G. Chen, W. He, J. Liu, S. Nath, L. Rigas, L. Xiao, and F. Zhao , ― Energy-aware server provisioning and load dispatching for connection-intensive internet services ,‖ 5 th USENIX Symp. on Net. Sys. Design & Implementation, pp. 337 -- 350 (2008) [1 9] R. Lock, T. Storer, I. Sommerville, and G. Baxter, "Responsibility Modelling for Risk Analysis," ESREL 2009 , pp. 1103 -- 1109 (2009)