A framework for learning about cloud computing

Profile picture for user pwainewright By Phil Wainewright December 6, 2013
What should universities be teaching students about cloud computing? Some answers were heard this week at IEEE CloudCom 2013

plug and cloud on sky, concept of cloud computing
Is the education system preparing enough individuals with the skills that industry needs to make the most of cloud computing? This was one of the questions raised among many during the IEEE CloudCom 2013 conference, which took place this week at the University of the West of England, near Bristol.

One way to answer the question is to respond with another: does it matter? Speaking in a panel on the final afternoon, Chris Swan of CohesiveFT said that it's inevitable when working in emerging technologies that new recruits don't arrive ready-primed with the knowledge they'll need. Instead, they should be adaptable enough to learn new skills as they go:

"It is a very dynamic industry and we need an education system that supports dynamic individuals."

The other way to answer the question is with a blunt no: not nearly enough. In fact, one speaker suggested that it is only with the advent of big data and a concomitant rise in funding for data science projects that cloud computing has finally become fashionable in academia. Like many of their IT colleagues in the commercial sector, academics had been prone to dismiss cloud computing as a passing fad not worthy of serious attention.

That is now changing — and indeed there are exceptions who are already well ahead of the game in their understanding of the potential of cloud. I was at the conference at the invitation of Dr Mohammed Odeh, who leads the software engineering group at UWE. He had organised an inaugural workshop on requirements engineering for cloud computing, which we co-chaired. I was there in my role as chair of EuroCloud UK.

Four principles

The need for a curriculum framework for cloud computing was addressed in a paper submitted to the workshop by Yuri Demchenko of the University of Amsterdam.

To the casual observer, designing a course curriculum may seem somewhat tangential to the topic of software requirements engineering. But it helped to emphasize a theme running through the workshop, that the success of a cloud computing project depends as much on non-functional requirements related to service provision as it does on meeting the functional requirements of the operational process itself. Making sure the necessary skills are present is a fundamental, albeit non-functional, requirement.

Demchenko outlined four core principles for effective professional education in cloud computing:

  • Provide knowledge both in Cloud Computing as a new technology and [in] background technologies
  • Empower future professionals with [the] ability to develop new knowledge and build stronger expertise; prepare [the] basis for new emerging technologies such as Big Data
  • Bloom's Taxonomy as a basis for defining learning targets and modules outcome[s]; provides a basis for knowledge testing and certification
  • Andragogy vs Pedagogy [adult not child teaching styles] as instructional methodology for professional education and training; course format [includes] on-campus education and training, online courses, self-study

He presented a way of visualizing the potential components of a curriculum, depending on the domain knowledge and skills to be transferred. The majority of users will only need to learn to work with virtual machines or applications in the cloud, he said.

Others will need a deeper set of knowledge to be able to manage a hybrid mix of cloud and on-premise assets: "You need a specific maturity level to integrate enterprise infrastructure with cloud."

The most advanced skills will be needed by those who work at the infrastructure layer of cloud providers. The full overview is represented in the diagram below:

demchenko-cloud-computing-curriculum-taxonomy 585px

Demchenko concluded with a discussion of potential course formats, including the question that relates to the elephant in the room of traditional institutions:

Will MOOC change the need for professional in-campus education?

His immediate response was that the importance of face-to-face interaction limited the value of online alternatives, but there wasn't time to delve into a debate on this. He also emphasized the balance of tutorial time with hands-on practical exercises.

Cloud learning

In my first visit to a university campus more or less since I left university myself, I was impressed by the vast number of Web terminals scattered all over the UWE campus that students are able to use on demand. But I saw little evidence that online access to information is complemented by the kind of online interaction that is being adopted by early adopting businesses today.

I suspect the academic world still underestimates the impact of cloud on its own business models and methods of operation, just as it has underestimated the wider significance of cloud computing. But that's not to single it out; most industries are making similar errors of judgment.

As a further aside, perhaps a vendor like Box ought to use some of its newly won venture funding to sponsor research into effective online collaboration strategies, as I suspect very few organizations are making the most of the potential. Certainly, most of the groups that I participate in, whether they use Box, Huddle or Google Apps, are all struggling to use them to maximum effect — even though by all accounts they're a great improvement on the likes of Sharepoint and Notes. Any findings could translate into better techniques for online education.

Getting back to the original topic, let me leave you with some food for thought on cloud learning. Here's a list of suggested questions from Demchenko, each designed to test a higher level of understanding of cloud according to Bloom's Taxonomy. How would you do?

  • [Knowledge] What are the main benefits of outsourcing a company’s IT services to cloud?
  • [Comprehension] Compare the business and operational models of private clouds and hybrid clouds.
  • [Application] Which cloud service model is best suited for a medium-size software development company, and why?
  • [Analysis] What cloud services are needed to support typical business processes of a web trading company? Give suggestions how these services can be implemented with PaaS or IaaS clouds. Provide references to support your statements.
  • [Synthesis] Describe the main steps and tasks for migrating IT services of an example company to clouds? What services and data can be moved to clouds and which will remain at the enterprise premises.
  • [Evaluation] Do you think that cloudification of the enterprise infrastructure creates benefits for enterprises, short term and long term?

Disclosure: Box is a diginomica partner. UWE funded the writer's modest T&E for this visit and he attended Cloudcom 2013 in his capacity as chair of EuroCloud UK.

Image credits: Sky© faithie - Fotolia.com; diagram courtesy of Yuri Demchenko