Skip to main content

“We have made a huge investment in the best breed of DevOps tools”, Sanjeev, CIO of a Service Organization, started sharing with me during my flight to Singapore. He continued, “Automation is key for us to drive DevOps. We are slow in adoption but optimistic about it for 2019”.

I became quite curious as DevOps is my passionate area. It was four hours flight, so I was happy that I got a companion to chat with. I started piling questions in my mind to ask him. I popped up the first question to him. “Sanjay, why you feel adoption is slow? What is stopping?”. With a smirk, he said, “I observed interesting patterns of failures with my teams.

[vc_empty_space height=”10px”]

Central DevOps team for CI/CD tool standardization

[vc_empty_space height=”10px”][vc_empty_space height=”10px”]

Automation does not equal to DevOps readiness

When we started the initiative most of our discussion were skewed over areas to automate or optimize the existing flow that would reduce cost and effort. We were able to introduce new tools and automated steps. We were unable to celebrate the work as overall efficiency improved a little. As we reflected on the initiative, we found the core issue were inherent in the architecture and approach the team created the unit and integration test suites. Automation had little to do with it.

[vc_empty_space height=”10px”]

Automation without feedback creates a vacuum

[vc_empty_space height=”10px”]

Tools are great to use. End to end flow had around 25+ tools. Even though we had orchestration challenges that hover over, we did not know if the tools we used are useful enough. In other words, the feedback loop was missing. It took time to trace the flow and understand the challenges. As the engineering teams used the tools we presumed the toolchain worked as expected. Interaction reduced and created a silo with the DevOps tools team. This caused Compliance and regulatory issues. Now we have Data-driven CI/CD process. The toolchain collects valuable data on every stage of the deployment pipeline. This data is used to gain insights and remediation steps to resolve any compliance issues.

I stopped asking more questions and started contemplating the challenges due to automation. This was a great insight to me.

[vc_empty_space height=”10px”]
[vc_empty_space height=”25px”]

About the Author

[vc_single_image image=”7869″]

Hari G, widely known for his keen analytical and product development skills, brings in 17 years of rich experience in execution of technology strategies and solutions that support company growth. As the CIO, Hari leads Agile, DevOps, Automation and analytics practices of TaUB Solutions.

Hari has extensive experience in user experience (UX/UI), Business Intelligence and Analytics, and building data-driven solutions using Agile methodologies

Next Post