Our Journey to DevOps: Which Came First … the Dog or the Dog Food?

Written by: Diana Hwang
3 min read
Stay connected

Editor’s Note: This guest blog post was written by Ari Livigni , senior principal software engineer, Red Hat, Inc.

I’ve worked at Red Hat for more than four years. When I started working here, we used many different technologies that weren’t our own products. In an effort to “eat our own dog food” as the saying goes, we decided to use Red Hat products and specifically the OpenShift platform to deliver our software. It’s been a great exercise in the customer experience, because it exposed some quality, usability and integration issues we didn’t realize our customers faced.

We suddenly saw the gaps in our product releases. The dog in our dog food strategy was showing us where we needed to make improvements rather than waiting for our customers to tell us what we needed to fix.

That realization led to our transition to DevOps. We began using continuous integration/continuous delivery (CI/CD) on OpenShift, an enterprise-grade application platform that’s built for containers with Kubernetes. The DevOps and CI/CD approach allowed us to address our issues quickly so we could deliver better and more robust products to our customers. This was our biggest success with DevOps. By eating our own dog food and moving to DevOps, we’ve been able to get ahead of our product development and release curve.

As you might expect, we’ve learned some lessons along the way. Here are a few key points:

  • There isn’t one tool that will work for everyone . You need to give your development team the freedom and flexibility to use the tools they prefer for maximum productivity.

  • Constantly iterate and improve the system . This will help alleviate fire drills and get your team into a reliable cadence of software releases.

  • Don't get distracted by the new shiny toy . There is always something new that grabs everyone’s attention for a while. Take the time to figure out what’s worth your attention and what’s hype.

  • Build monitoring and metrics into your system . This information helps you and your team understand whether the solution is working as intended, and it also gives you great data to showcase the value-add of DevOps to your management team.

  • Don't be afraid to fail . It’s part of the DevOps journey. Failure is difficult, but it’s so important. It’s where you regroup and learn from your mistakes.

These are just a few details about our successes and insights as we’ve embraced DevOps and CI/CD. At the DevOps World | Jenkins World 2018 conference , September 16-19, I’ll be speaking on a panel called “Bumps, Bruises and Bandaids: Real-World Lessons from Practitioners” with several of my peers from Medallia, Intuit , Electronic Arts and American Airlines. Be sure to attend this session to hear more about our ups and downs, successes and failures, as well as organizational impacts and points of growth with DevOps.

Stay up to date

We'll never share your email address and you can opt out at any time, we promise.