Cloud Education Lags at Universities: Four Professors' Perspectives

Cloud computing is a wildly in-demand skill that is still not taught at many institutions of higher education. We spoke with four professors of cloud computing to learn how they approach teaching the topic, the challenges they face, and the benefits of arming graduating students with practical cloud computing skills.

In October, LinkedIn unveiled their list of the “top skills that can get you hired in 2017.” For the second year in a row, cloud and distributed computing led the list.

How do you learn cloud computing skills, though, given the technology’s relative youth? Cloud computing as we know it today only experienced explosive growth in the past fiveLinkedIn Top Skills 2016 years.

We spoke with four professors of cloud computing from Cornell University, Georgetown University, Carnegie Mellon University, and the University of North Florida to explore the state of cloud computing education at institutions of higher education.

Through the interviews, we explore what cloud computing courses look like at a college level, the challenges professors face when designing and teaching cloud computing, and the value a cloud computing course can offer a student.

Though the four professors we interviewed have all taught courses on cloud computing, they are in the minority. Many universities still do not offer courses on cloud and distributed computing, despite the growing need for these types of skills in the workforce.

By not investing in cloud computing education, institutes of higher education are potentially inhibiting the technical development and job potential of a large swath of qualified students.

What does a cloud computing course look like?

There is no shared framework or generally accepted structure for cloud computing courses at universities, primarily due to the topic’s youth. For example, Amazon Web Services, the dominant cloud computing provider by a wide margin, only launched a little over a decade ago in 2006.

Compounding that issue is the sheer breadth of the subject, a fact mentioned by many of the interviewed professors. Despite its short history, cloud computing is ripe for in-depth study, with possible topics ranging from architecture, to data management, to social and legal impacts.

Thus, given cloud computing’s recent emergence and complexity, how should a course be designed?

The answer to this question likely comes down to the goals of the university, as well as the faculty’s available knowledge.

The four courses we examined were offered through computer science and software engineering departments and mainly focused on architecture and the driving technologies behind cloud computing.

Cloud Computing at the University of North Florida

The University of North Florida (UNF) offers “CEN 6086: Cloud Computing,” a graduate-level course with a significant research component.

UNF first implemented CEN 6086 in 2013, when Dr. Sanjay Ahuja received a summer teaching grant to develop a cloud computing course.

Dr. Ahuja is a Full Professor of Computer Science at UNF and received his Ph.D. in Computer Science and Engineering from the University of Louisville. He was the FIS Distinguished Professor in Computer and Information Sciences from 2010-2014 and directs UNF’s Cloud Computing Research Group (CCRG).University of North Florida

Dr. Ahuja’s course is popular. It has been taught two more times since its introduction – in 2015 and 2017 – and attracts many students, with over 60 graduate students completing the course since its inception.

The course is a blend of theoretical and practical education. It’s “a hybrid of research-driven content and work with public clouds, specifically Amazon EC2 and Google Compute Engine,” said Dr. Ahuja.

Some of the topics covered by the course include:

  • Cloud computing models (Infrastructure/Platform/Software-as-a-Service)
  • Public, private, and hybrid clouds
  • Data centers for cloud computing
  • Cloud Architectures and Performance
  • Principles of virtualization platforms
  • Security and privacy issues in the cloud

Students are graded based on their performance in a final exam, homework assignments, presentations, and a term project focused specifically on scientific research in cloud computing.

Cloud Computing at Georgetown University

Georgetown University regularly offers a cloud computing course as an elective for the Technology Management Master’s program in the School of Continuing Studies. It is taught by Kevin McDonald.

McDonald is not a career academic. He serves as the Founder and Managing Director of GreyStaff Group, LLC, a cyber security strategy and solutions firm, alongside teaching at Georgetown UniversityGeorgetown as an adjunct faculty member.

Georgetown’s course is divided into three parts:

  • Part 1 covers the basics of cloud computing, such as infrastructure, how a data center is run, and what issues and concepts are commonly encountered.
  • Part 2 discusses the terminology of the cloud, how it translates to legacy data centers, and how to transition from a legacy data center to a cloud system.
  • Part 3 encourages students to apply their knowledge in a real-life practicum where they are required to deploy a mobile app using a cloud-based prototyping software.

The details of a cloud computing course can be specifically swayed by the university’s specialties.

Part two of Georgetown’s course is influenced by the university’s proximity to the federal government in Washington, DC, according to McDonald. There is incentive within the government to use cloud computing techniques and understanding how to transition to cloud from a legacy system is a valuable skill.

Cloud Computing at Carnegie Mellon University

Carnegie Mellon University offers a project-based online course on cloud computing for students at their campuses worldwide. Students complete a project every week, building solutions on Amazon Web Services, Microsoft Azure, and Google Cloud Platform.

The course was introduced in 2013 and its developers included Dr. Majd Sakr, a Professor of Computer Science at Carnegie Mellon, as well as three engineers and a postdoctoral researcher. The team published research evaluating the course’s development, outcomes, and obstacles in 2015.

Overall, the 2015 research recorded favorable feedback from students on the course, as well as high course completion rates. However, the authors also identified several pain points, including:

  • The cost of using Amazon Web Services for projects
  • The large time investment required to build and test such a course
  • The difficulties of keeping students engaged with timely fine-grained feedback, despite the distance inherent in an online course

Cost as a pain point is investigated further in the next section of this article. If you are interested in reading more of Dr. Sakr and his colleagues’ research, you can find it below.

Now in its ninth offering, the course has students from Carnegie Mellon's Pittsburgh, Silicon Valley, Australia, and Rwanda campuses. The course staff includes the professor, two full-time staff members, and two-dozen teaching assistants. Dr. Sakr is working on offering the course to industry professionals needing training on the cloud to expand its reach to non-Carnegie Mellon students.

What are the challenges of developing and implementing a cloud computing course?

The four professors of cloud computing we interviewed identified three main challenges of developing and implementing a cloud computing course at an institute of higher education: cost, the ever-changing nature of the topic, and limited on-campus expertise.

1. Cloud Computing Courses Can Rack Up High Price TagsChallenge #1 - High Price Tags

Cloud computing courses can potentially be cost prohibitive at the university level. Dr. Sakr’s research at Carnegie Mellon found that, “During the first three offerings, the course cost approximately $300, on average, per student.”

The higher cost of a cloud computing course is a result of the cost of using services, such as Amazon Web Services, in hands-on projects. Cloud computing platforms normally charge based on usage, with greater usage of the platform’s services leading to a higher bill.

Dr. Sakr mentioned that “negligent students” led to runaway costs, as they did not pay close attention to the breadth of their usage and thus their projects’ eventual cost.

To incentivize students to reduce costs, Dr. Sakr and his team made the budgetary aspects of cloud computing part of the students’ final grades.

“If [the students] go above the budget for a project, they get a 10% penalty [on their grade]. If they go above double the budget, they get 100% penalty,” explained Dr. Sakr. “So, you can explore and make a couple of mistakes and still get a 90, but if you're going to be a little bit more careless, then you will pay significantly.”

Dr. Sakr found this method to be highly effective at controlling runaway costs of his students’ projects.

Furthermore, working within a budget is a valuable skill for future cloud computing professionals to gain. “What we've heard from our industry partners is that it would be wonderful if the students are going to program this very large cloud, that they also can be thoughtful about the budget,” Dr. Sakr said.

Thus, a potentially serious obstacle of a cloud computing course – working within a reasonable budget – becomes a boon for its students. Cloud computing jobs do not exist within a vacuum. Businesses have budgets for their technology, so it is important for students to learn how to manage that aspect of a potential future job.

2. Cloud Computing Is Still Constantly Evolving

Another challenge is teaching a field that is still experiencing constant and brisk innovation. One professor, Dr. Ken Birman of Cornell University, even goes so far as to advise universities against implementing cloud computing courses until the field slows down its advancement.Challenge #2 - Constant Change

“I do not think any university is able to teach this topic because of its lightning fast evolution,” said Dr. Birman. “As a purely pragmatic matter, we cannot teach the area until it begins to slow down and hold still for at least a few years at a time.”

Cloud computing technology is undergoing constant change. For example, consider the fact that Amazon Web Services released 1,017 total updates in 2016 alone.

Birman elaborated on the challenges of cloud computing’s constant evolution by explaining the learning curve involved in understanding a new product.

“Often, [new products] might individually take a few weeks to learn to use. So, to teach a topic, you may have to spend weeks learning the technology, then developing a lecture, and yet the content would be stale a year later.”

Academia’s traditional timeline of in-depth research and development can run against the accelerating pace of technological innovation. This is an issue that universities may face in a number of fields in the future, as the speed of technological innovation continues to increase.

The question is, should universities let the fact that technology advances quickly – and most likely will continue to advance at an increasing pace – inhibit/control the courses they teach their students?

The fact is that this quickening pace may never slow down again, according to Ray Kurzweil, a scholar and futurist who has written a book on the acceleration of technological advancement. “We won't experience 100 years of progress in the 21st century — it will be more like 20,000 years of progress (at today's rate),” he wrote in 2005.

Given that the increasing pace of technological innovation may be the new normal, how will universities adapt? How can they revise traditional scholarly endeavors to meet new speeds of invention?

These questions perhaps cannot be answered now but require serious consideration in the future.

3. There is a Lack of Available Expertise

A third challenge is finding faculty with the skills, experience, and commitment to teach students cloud computing and keep up with the changing industry.Challenge #3 - Lack of Expertise

Not having qualified staff members to help facilitate the class is a serious issue at Cornell, according to Dr. Birman: “We have no experienced staff to do grading or run recitation sections, since so much is new.”

While the issue is not as serious as at Cornell, Dr. Ahuja of UNF specifically mentioned staff expertise when asked to list the biggest obstacles of teaching cloud computing.

“In terms of resources, a knowledgeable professor, grad student, or assistant is necessary. Someone who is very educated about cloud computing and architecture, including platforms like Hadoop, and automation tools, like Chef and Puppet.”

This type of academic may be difficult to hire for a university, however, given the cloud computing industry’s relative youth.

Dr. Ahuja suggests that, if a university cannot hire a faculty member with expertise in the cloud, they should offer training for existing faculty who are curious about the topic, or hire industry leaders with an interest in teaching.

What are the benefits of teaching cloud computing?

Cloud computing skills are in-demand in the workforce, and universities can give students an incredibly powerful advantage by offering cloud computing courses that build technical skillsets.

A late 2012 research report found that 1.7 million cloud-related positions remained unfulfilled worldwide due to a lack of necessary skills and training. While more recent research of this type does not yet exist, it is clear that cloud computing expertise is still highly desired.

Dr. Ahuja, Dr. Sakr, and McDonald were adamant in their support of implementing cloud computing courses because of the topic’s increasing omnipresence in today’s technological landscape.

“It’s becoming more important to understand cloud computing simply because … it’s being adopted quite rapidly now,” said McDonald of Georgetown University. “Having in-depth experience and knowledge of the cloud is probably a core competency going forward.”

Kevin McDonald Quote Card

As more and more companies transition away from legacy data centers and adopt cloud computing, these skills will only increase in demand.

Dr. Sakr emphasized the advantage of having cloud computing skills when job hunting. “Our students are very competitive when they are interviewing with leading companies,” he said.

Many companies are specifically searching for cloud expertise. For example, Computerworld’s Tech Forecast 2017 found that over one-quarter of US enterprises plan to hire for cloud computing skills in 2017.

Dr. Ahuja spoke similarly, stressing the value of the knowledge cloud computing courses impart on their students.

“My students have become experts in launching and running virtual machines, as well as writing and running scripts on them,” he said. “These are marketable skills which they can talk about when interviewing for a position.”

In tough job markets, cloud computing may be the skill that pushes a candidate over the edge.

Cloud Computing Is Here to Stay

All signs point to a continual rise in cloud computing usage among companies for the near future. Gartner predicts that the worldwide public cloud services market will increase by 18% in 2017.

With that trend in mind, a number of universities have begun offering opportunities for learning cloud computing at their institutions. This includes the University of North Florida, Georgetown University, and Carnegie Mellon University, where courses have been implemented with relative success.

However, for many universities, teaching cloud computing remains out of reach due to higher costs, the technology’s rapid pace of innovation, and a lack of available expertise.

At Cornell University, these obstacles significantly affected the success of their broad cloud computing survey course. The story is likely the same at many other universities or colleges.

The decision whether to pursue the development of a cloud computing course comes down to the specific needs of the student body, available faculty, and the department’s unique goals, among other factors.

However, as cloud computing continues to increase in popularity, it may be worthwhile for most institutes of higher education to at least consider the topic for future implementation in their curricula.

About the Author

Riley Panko, Content Developer and Marketer at Clutch

Riley Panko is a content developer and marketer at Clutch, a B2B research firm in the heart of Washington, DC. Her research focuses on the cloud. Reach out with questions, comments, or concerns at [email protected].

 


More Research on Cloud

  1. Six Tips for Expanding Your Cloud Security Knowledge
  2. iCloud Security Survey 2017: Is It Safe?
  3. Cloud Storage and Security: The Rundown

Subscribe to Clutch’s Newsletter