10 Ways to Transition Traditional IT Talent to Cloud Talent
While many IT professionals love learning new things, IT leaders and their organizations must do several things to facilitate a smooth transition to cloud.
![](https://eu-images.contentstack.com/v3/assets/blt69509c9116440be8/bltf3b1def83b9500ed/64cb56875f98791e867062c1/000-skills-pixabay-cp.jpg?width=700&auto=webp&quality=80&disable=upscale)
Organizations continue to transition more storage, compute and applications to the cloud. However, to succeed, they need the right skills and processes in place.
"I grew up in the traditional IT world. Things were simple then. If you knew a few technologies you could really make a difference, but in the cloud there's a lot more interconnection between the different technologies and a lot more options available," said Neeta Jain, CTO of Vibrent Health. "The complexity is very hard, so the learning curve is higher than in the past."
Jain and others interviewed for feature said that traditional IT and cloud have a lot in common because even with cloud, companies still need systems, network, data and security expertise. It's just that fewer bodies are needed to manage and maintain infrastructure than when everything resided in an on-premises data center.
Today's greater dependence on cloud isn't always matched by the right mix of skills within organizations, however. As a result, some businesses have learned the hard way how to optimize their use of cloud services and the associated costs. Cloud security also tends to be an issue.
Quite often, when organizations migrate to cloud, they assume they need to hire all new cloud talent to make it happen. However, existing talent can be upskilled to help fill the gaps. With that in mind, many companies are providing in-house training or paying for third-party or curated third-party programs. However, training programs and technology need to be supported by updated processes that are more in line with the company's shift to cloud-first architecture.
"You need your existing staff, so make sure they have the training they need. The second part is about trust or autonomy. You need to trust these trained engineers will do the right thing," said Dave Anderson, director of technology at insurance company Liberty Mutual. "We [also put] guardrails in place so our engineers can try things and experiment in a safe way."
While many IT professionals love learning new things, organizations must do several things to facilitate a smooth transition to cloud. Following are a few tips.
Organizations continue to need more cloud skills as more compute, storage, data and applications move to cloud. For some companies, DevOps has fueled the transition because new applications are being built as cloud applications. Insurance company Liberty Mutual started exploring DevOps several years ago. In doing so, it transitioned the traditionally siloed waterfall teams into cross-functional DevOps teams.
"The three things we have focused on are digital transformation, becoming more customer centric and moving to the cloud. When you take all three of those things, they all point to DevOps," said Dave Anderson, director of technology at Liberty Mutual. "You've got the autonomy to do what you need to do. Instead of being handed a piece of work, you're handed a business problem and that's extremely empowering for an Agile team."
One of the biggest complaints that IT leaders have about moving to the cloud is a lack of cloud expertise. In fact, team health is less defined by roles today than the mix of skills team members collectively possess. If you don't know what skills you already have, you don't know what you need. And you may not know what you have if there's no way of inventorying skills. Liberty Mutual tracks self-development, certifications and collaboration across different teams, among other things, to ensure its IT staff is moving in the right direction.
"I think certifications have a lot of value, but it's worth canvassing a team because many people have side projects or like to work on things at home in their own time. Maybe they've hopped around three or four jobs in the past few years so they have a lot of varied experience that you might not see in your day-to-day operations," said Jesse Stockall, chief architect, cloud management at technology intelligence platform provider Snow Software. "Unless you're a hiring manager who still has their CV, you probably don't know every tool or every piece of tech they work with."
Enterprise content and process management software provider Hyland Software has undertaken significant career mapping initiative for the roles required today and tomorrow. The company has backed up that initiative with various training resources employees can access to develop new skills. According to Senior Technical Recruiter Christina Gibson, many traditional systems administrators and DBAs have moved int DevOps roles.
"We try to make sure that individuals are well informed of what opportunities there are within the organization, whether it's through our career development initiative, performance development and conversations with their direct managers," said Gibson. "We've made our training programs as accessible to everyone as possible whether it's in-person classes in the office, online learning, or websites and tools such Udemy, things that are even outside of Hyland that employees can research and discover."
Skills mapping needs to be tied to career path development. For example, Vibrent Health's systems admins moved to DevOps, then became site reliability engineers (SREs) or maintenance reliability engineers (MREs). Now they work in DevSecOps, according to CTO Neeta Jain.
Financial industry technology solutions provider Broadridge Financial Solutions recently shifted its site reliability engineers (SREs) left because they had already been working closely with developers to build automated pipelines which include automated deploys. According to senior technology fellow Mark Schlesinger, the infrastructure teams are still responsible for the infrastructure that's common across applications and the SREs are responsible for ensuring that the infrastructure for an application or set of applications is appropriate.
Sounil Yu, CISO-in-residence at YL Ventures was the Chief Security Scientist at Bank of America for more than 30 years. While there, the company was transitioning a traditional IT environment to one that was more cloud native. One of the biggest differences between the two environments is the interface, Yu said. For traditional IT environments, point-and-click interfaces work well for designing, building and management systems, but cloud architectures require coding skills.
"In a cloud environment the interface is an API [because] cloud environments are largely software-defined. With more and more enterprises moving to fully software-defined data centers and IT environments, it has become critical for IT staff to understand how to program to be able to define and manipulate this environment. To that end, learning languages like Python, which is part scripting, part full-fledged language, can serve as a way to upskill IT staff."
Developers know how to code, but they're not infrastructure experts. Conversely, infrastructure experts are not programmers. One way to combine two types of expertise is for IT to create infrastructure or architectural patterns for developers. At Liberty Mutual, that might mean creating an image in the cloud image or an API gateway, for example.
Although the patterns help bridge the two disciplines, they are not a substitute for actual systems, applications, networks, communications, security and data expertise as is evident by the number of misconfigurations in the cloud. YL Venture's Yu said the specialists in those areas should ensure developers are doing the configurations properly and that everyone should be speaking the same language, which is oftentimes expressed as API and programming code.
In a cloud environment, consistency and repeatability are critical at scale. If everything is configured manually, people will make mistakes that could be eliminated by automation. Automation is also critical for DevOps teams, especially those that have advanced to CI/CD because time to market and product quality cannot be achieved simultaneously if everything from coding to testing to deployment must be done manually.
"The best people constantly automate themselves out of a job. It's a strange metric, but how often are they taking something they would otherwise do manually and codify [it]?," said YL Venture's Yu. "In a traditional IT environment, you might have one person that can handle 100 or 1,000 systems, but today you should easily be able to scale to 10,000 or 100,000 systems."
Transitioning from traditional IT to cloud affects the tech stack, IT processes and people. Many interviewed for this slide show explicitly stated that change should mutually benefit the company and its employees. Celebrating employees' accomplishments, providing company-sponsored training, soliciting feedback, giving employees autonomy and compensating IT staff fairly help motivate staff. Not surprisingly, those things also help a company attract and retain talent. However, processes also need to change as staffing evolves.
"[Changing] how business processes work is sometimes a little harder than just getting somebody trained up on a particular skill," said YL Ventures' Yu. "The business processes have to be aligned with the roles and responsibilities people have.”
Deploying infrastructure with an application is efficient, but it can also be expensive if not managed well. Cloud cost control is a matter of education, but it's also a function of processes.
"Cloud costs can run away from you very quickly, so a big part of optimization is making sure that you've optimized for cost controls as well," said Rich Murr, CIO of cloud software provider Epicor. "You really have to understand how your footprint leverages cloud resources. It's almost a full-time job."
Cloud services are proliferating so rapidly that it's impossible to keep up with everything. Part of managing a transition to cloud is to stay focused on what the company is trying to achieve rather than getting distracted by sexy stuff such as robots which may have no value to your organization in the foreseeable future.
"It's easy to get lost in the proliferation of cloud services but probably most of the time, your compute, storage, security and networking services are used to compose almost everything," said Epicor's Murr. If you get good at those, then you can start to figure out what else your business might be able to use."
Check out other InformationWeek slideshows.
Cloud services are proliferating so rapidly that it's impossible to keep up with everything. Part of managing a transition to cloud is to stay focused on what the company is trying to achieve rather than getting distracted by sexy stuff such as robots which may have no value to your organization in the foreseeable future.
"It's easy to get lost in the proliferation of cloud services but probably most of the time, your compute, storage, security and networking services are used to compose almost everything," said Epicor's Murr. If you get good at those, then you can start to figure out what else your business might be able to use."
Check out other InformationWeek slideshows.
-
About the Author(s)
You May Also Like