Behind the scenes: A day in the life of a CloudOps director

3 years ago 301

The key to adopting a smart DevOps mindset, according to a CloudOps director? Prioritization and automation.

Nelson Normahomed, born and raised in the small city of Mutare, Zimbabwe, first got into technology with the arrival of his first computer, around age 15. "I was loading up, installing Windows, and just doing the basics, but my interest grew from there," he said.

Now, Normahomed is the senior director of cloud operations and IT at the software company Vendavo. In his nearly five-year tenure, he's taken a lead in helping adopt a DevOps mindset and best practices at the company.

His formal path began when Normahomed moved to London, England, and earned a degree in computer systems and networking, getting side jobs in freelance web development and computer repairs to get by. When he graduated, he took a job at Fujitsu—his entry into IT—where he did first-line, second-line support, dealing with users' issues with applications, servers and more. Later, he got a job as system engineer for a healthcare company based in Kansas City, where he started doing infrastructure implementation, and moved into a technical architect position, and stayed at the company nearly a decade.

When he first began at Vendavo, Normahomed was a technology architect, managing a data center and getting customers to migrate to it. Then, the company transitioned from a managed data center to public cloud providers like Microsoft and AWS. His role was both as technology and cyber reliability engineer—"to design what our new cloud would look like." As a system cyber reliability engineer, he would take an active role in implementation—installing the cloud system, building out, and running the solutions—versus simply handing it over to customers.

SEE: Research: Video conferencing tools and cloud-based solutions dominate digital workspaces; VPN and VDI less popular with SMBs (TechRepublic Premium)

Since Vendavo owns the cloud it supports production environments, in first-line, second-line fashion; however, "the whole cloud infrastructure supporting that application remains with us, our team, and we typically support that moving forward," Normahomed said.

"Every day is a new challenge," Normahomed said, since Vendavo is constantly adopting the latest technology.

"We don't stagnate, technology-wise, because we're always trying to keep up with what makes sense from a technology standpoint and incorporate that into our infrastructure appropriately," he said. "Everything from infrastructure as code, working with Microsoft and AWS cloud providers, Oracle and Microsoft SQL server—we touch it all."

On a typical day, Normahomed starts his schedule of tasks around 9 a.m., "but in the life of an SRE, that always changes, right?"

The main goal is to keep the customer's infrastructure running reliably, he said, and make sure "there are no surprises." To do this, he works closely with engineers and project teams. The next priority, after the customer, is project work—such as getting code implemented into the customer's environment—followed by internal development tasks for the cloud.

And part of the day-to-day work is about improving infrastructure. This means automation. "How much can we automate? Because that's pretty much our main priority: performance, reliability and automation."

SEE: Quick glossary: DevOps (TechRepublic Premium)

Anything they can automate, he said, they will, to reduce work on mundane tasks and focus on critical issues.

"So, we are constantly observing and monitoring the infrastructure," he said, which includes determining how to respond to hundreds of alerts.

DevOps has been around for about a decade, but Vendavo's journey is about a year and a half old.

"We went down this path because, initially, we were very siloed as organizations," he said. The cloud operations team and engineering team did not always "have full visibility in terms of how we build out our infrastructure within the cloud or how we run operations within the cloud, supporting our customers," he said, which posed a challenge. In a smaller development environment, there was also a services team, focused on new features or upgrading code—but there wasn't a lot of visibility about how they operated in the production environment. This could lead to some shifting of blame, or there could be difficulty securing an engineer's help in troubleshooting an issue.

SEE: AWS Lambda, a serverless computing framework: A cheat sheet (free PDF) (TechRepublic)

By implementing DevOps, "we can have a fluid pipeline of code delivery, accountability and support, and a whole loop of feedback," he said. "From the time it gets deployed, or from the time the Engineering team propagate that code into the production environment where the customers were using it, all through to troubleshooting, and having that feedback back to them, so they can also understand and know the issues and the challenges that we faced by then also making them understand how we are fully set up in the cloud, as well, from an infrastructure perspective."

The DevOps roadmap includes engineers from cloud operations and cyber reliability engineers from the CloudOps team, who are plugged into Oracle. Together, they have weekly stand-ups and sometimes daily stand-ups. This has already eliminated the silo.

The communication also helps by prioritizing challenges instead of simply responding to JIRA tickets, which could cause low resolution times.

The main skills needed for the job? The ability to prioritize and automate, Normahomed said.

"As an engineer, the exciting part is dealing with challenges, troubleshooting and resolving issues," he said. "And there is no limit in terms of what technology, or how much, you can automate."