EPAM Systems’ 2023 Cloud Mastery Report shares strategies for engineers.
Engineers start to master the cloud when they work with managers that set cloud-related goals for themselves and their organizations, according to EPAM Systems’ 2023 Cloud Mastery report, titled “From Taming Cloud Complexity to Achieving Cloud Mastery.” EPAM’s survey of over 400 senior Information Technology leaders in the U.S. and the U.K. shows that skilled use of the cloud requires mentorship, practice and determining what tools pair best with the cloud.
Cloud mastery allows engineers to build and deploy applications more quickly, efficiently and securely. It also makes it easier to crunch and store data used in AI and machine learning (ML) algorithms, simulations and other engineering workflows. But it easy to implement.
“Just completing certification programs and updating the course catalog won’t do it,” says Sandra Loughlin, chief learning scientist and global head of Client Learning & Enablement at EPAM. “Engineers and managers need to find out what use of the cloud looks like in practice.”
Switching to technologies like coding language, compilers and source controls that are cloud-friendly or cloud native speed up the process. Then engineers can build or utilize apps that take full advantage of the flexibility of cloud-scaling. They can also better work with the elasticity and resiliency that the cloud offers.
“You want software engineers to be able to write high-quality code with a low amount of errors and exceptions,” says Miha Kralj, vice president of Cloud Advisory and Transformation for EPAM.
EPAM’s report found few differences between workers’ use of the cloud between the U.S. and the U.K. Instead, there was a greater difference between industries. Workers in high-tech and communications tended to push for cloud mastery when it offered them speed, agility and access to new innovation. In contrast, workers in the public sector, like governments, tended to be very risk averse. They used the cloud primarily when it offered cost and risk reduction.
EPAM is a Newton, Pennsylvania-based company that specializes in digital platform engineering. It created the report to better understand what it takes for employees and organizations to work well with the cloud. Respondents to the survey included C-level, VP-level, and Director-level employees in industries ranging from healthcare to financial services.
Widespread Cloud Adoption, but More Training Is Needed
The report indicated cloud adoption is strong. Almost all—97 percent of the organizations surveyed—had a cloud center of excellence. Yet only 34 percent of the organizations had the necessary cloud skills in-house. Approximately a quarter of organizations, 26 percent, had to lay off employees to make room for people with the right skills. Close to a third, 29 percent, had trouble hiring skilled talent and had to pay a premium for it.
Most respondents, 62 percent, said their first reaction to the term “cloud mastery” was that the term was about technology. Only 15 percent of respondents said their first reaction was that cloud mastery was about business.
It is impossible to make a strict comparison of data center and cloud costs. Data centers, servers, storage, networks and software licenses are depreciated capital expenses with perpetual rights. Organizations often use them long after they write off the costs. Cloud costs are usually calculated with a pay-as-you-go model, since costs are consumption-based.
Yet a switch to cloud computing typically helps organizations reduce IT costs. Cloud computing is particularly useful for organizations that want to reduce their IT budget or are uncertain about their future IT needs.
Kralj says the cloud is the only place where data storage and computing resources can be added or removed with a simple application programming interface (API) call.
“[This removes] the typical bottleneck in traditional environments. Provisioning time for additional compute and storage used to take six months or more,” says Kralj.
In the past decade, organizations have evolved to see the cloud as more than a cost-cutting “IT thing.” Now they see it as a driver to develop new business models, markets and revenue streams. Approximately 77 percent of respondents thought business transformation was a key benefit of the cloud.
Respondents showed interest in using technologies in the cloud to drive IT and business innovation. In particular, organizations wanted to be able to adapt their projects to Web3, the new iteration of the World Wide Web, and the metaverse.
One of the most significant ways that the cloud helps engineers is by resolving concerns that affect network service quality and operations. A cloud with functionality that involves a great deal of automation makes it easier for engineers to complete projects. This is because the engineers deal with fewer disruptions and manual interventions.
Organizations that use the cloud instead of on-premises infrastructure can see IT jobs
transform from blue collar into white collar. This is because many routine tasks are automated in the cloud. A shift to the cloud usually involves the IT team changing its priorities from problem management to proactively testing the cloud.
Kralj describes the work of service reliability engineering and chaos engineering as trying to cause “tiny earthquakes.”
“The IT team attempts to shut down the cloud services underneath the application,” says Kralj. “It tests the application’s reliability. When the team makes the application stronger, it can become more self-healing. This facilitates the jobs of other professionals, including engineers.”
The improvements free up time for engineers to complete their educations relating to the cloud.
According to Loughlin, all levels of engineers should start with a foundation of cloud training and certification programs. These should be augmented with hands-on experience and mentorship from more experienced peers. Ongoing learning opportunities like webinars and conferences are beneficial.
Organizations should also provide engineers with a library of evergreen reference material and space to experiment with Infrastructure as a Code (IaaC) and automation.
Working with an experienced, trusted partner on cloud training can accelerate an organization’s adoption and mastery of the cloud. Not taking advantage of partners’ expertise can lead to business trouble.
“We’re talking about partners [that] have migrated applications to the cloud, [that] deeply understand cloud security and compliance [and] can ensure the highest efficiency and cost-effectiveness,” says Kralj. “Organizations utilize partners in all aspects of their cloud strategy, development and operations.”
Loughlin adds truly mastering cloud upskilling, communication and change requires support that is inclusive.
“[This is] to address the whole-organizational transformation imperative,” says Loughlin. “Employees [shouldn’t] feel side-lined, ignored or pressured. Learning cloud skills should be seen as a positive motion that improves people’s work, career and future.”
Future Reports: What Will Be Included?
EPAM’s next reports on cloud mastery will involve longitudinal studies. The goal is to see how organizations shift their use of the cloud over time. EPAM plans to survey organizations beyond the U.S. and the U.K., particularly in several “geopolitical basins” that share features like a common language. EPAM wants to hear from respondents in a wider variety of industries to learn how goals and uses of the cloud differ.
EPAM also plans to add questions to assess how the adoption of large language models (LLMs) is impacting enterprise cloud ecosystems. Currently, LLMs and generative AI are not aware of what is true and what is false. Those decisions are still left to people.
“We need to oversee how to include such tools in the cloud to make sure AI is used ethically,” says Kralj. “That way organizations can do their job properly.”
For example, an AI-powered performance assessment tool might focus on the number of lines of code an employee wrote in a day. The best developers generate negative lines of code. This is because they optimize and cut waste from the codebase.
“The goal is shorter, cleaner and smarter code. Right now a person is still the best judge of the quality of an employee’s work,” says Kralj.
