Whilst taking advantage of cloud services can be a fairly painless process that rapidly delivers benefits for small and medium-sized organisations, education institutions may require a more thorough risk assessment and planning process to ensure they are ready to make the move and able to deliver real benefits.
Having said that, the sector’s experience with grid computing (joining up disparate resources to increase processing power for research purposes) and federated access management (FAM) to control access to resources across a range of different organisations, will stand it in good stead for getting to grips with these new developments.
Opinions differ as to the best way for the education sector to move into the cloud. In the Netherlands SURF (2011) takes the view that:
"Institutions would only need to develop private solutions if no suitable public or community cloud services were available or could be developed."
Indeed they would go further than this and put the decision-making authority firmly in the hands of the end users:
"Give end users the freedom to choose between public cloud providers. Ensure that the presence of a good infrastructure means that it does not matter what they in fact choose."
In the UK the approach tends to be a little more cautious. What is common in both the UK and Netherlands is that few institutions have yet developed formal strategies and policies on the use of cloud computing resulting in the risk that practice will outpace policy in this area. There is anecdotal evidence (see for example the Kindura Project case study) that many researchers are already bypassing institutional IT departments and taking advantage of cloud services in order to deliver additional storage capacity very rapidly.
Such a piecemeal approach to contracting for cloud services is unlikely to deliver genuine economies of scale and will make it difficult to provide good governance in order to ensure effective data management and security. Our detailed guide on enterprise architecture outlines how an organisation may develop a roadmap for technical developments ensuring best fit with vision, strategy and business process.
One aspect of the decision-making process around moving into the cloud is to consider the nature of the information to be hosted and or processed in a cloud service. Another is to look at the technical readiness of the organisation.
Information issues – The cloud computing toolkit produced by Aberystwyth University for the Archives and Records Association (Convery 2010) is written from the point of view of an information professional and offers comprehensive guidance on assessing the risks of migrating certain types of information to the cloud. Such decisions need to be made in the context of a legal and regulatory framework that is still evolving and the reader is referred to the section on Legal Issues in the Cloud for details of up to date reference sources.
Technical issues – The FEAST report (Clark et al 2011) suggests an incremental approach to developing experience and capability in the cloud environment starting with familiarity with server virtualisation and moving into private cloud space as a precursor to hybrid and public environments.
"Migrating towards the public cloud environment may be best facilitated by building a private cloud and over time building a hybrid cloud with a commercial partner(s) and fully understanding the cost-models, the contractual requirements, etc in order to adequately determine a return on investment (ROI) and risk assessment as parts of the business case."
Virtualisation has its roots in mainframe computing but has been a key enabler in the development of cloud services. Virtualisation permits the concentration of large amounts of storage and processing capacity in ‘server farms’. Data relating to different systems (or even different organisations) can be hosted on the same server but kept virtually separate. This eliminates the need for lots of separate physical servers all with redundant capacity for scalability and backup purposes.
The decision as to which cloud deployment model to explore in the first instance will, to some extent relate to the type of service being outsourced to this environment. There may be some cases where activities that are not mission-critical and do not handle sensitive information can be piloted straightaway in the public cloud. The Kindura Project created a research repository using a hybrid approach recognising that this involved additional complexity and cost but feeling that this was justified given the current state of the market with regard to matters such as legal issues and technical standards.
The following diagram shows a roadmap for moving towards use of the cloud for research purposes from the Jisc TeciRes (technical review of using cloud for research) project (Wills et al 2010)
"The cloud is generally not for beginners; it is wise to have considerable experience with a virtualised server environment before migrating services towards the cloud from this environment."
Clark et al 2011
"Typically institutions show a great deal of interest in cloud computing although there is a range of levels of involvement. … Despite this considerable interest and more limited use, the review revealed that few institutions have formal polices on adoption or use of cloud computing."
Macdonald et al 2010
"A large range of public cloud services are available. Students and staff are already using them but in a fragmented and non-cohesive manner. And although only a few institutions have drawn up their own strategy for cloud services, they are nevertheless all on the verge of making the relevant choices. It is by no means inconceivable that the lack of a strategy will lead to those choices being hasty and insufficiently well thought out."
"The hybrid model results in additional costs in managing both internal and external resources and in determining which content may be moved to external cloud providers. Given the current state of the cloud infrastructure market, we believe that these costs are justified in the short to medium term. However, in the longer term, it may prove to be feasible to adopt a purely outsourced storage model."
Kindura Project 2011