CASE STUDY: The Getty Achieves Continuous Delivery Agility in Cloud Native App Deployment on AWS
The Getty, based in Los Angeles, pursues a global mission to support further understanding of the visual arts. That mission relies on The Getty’s ability to support digital image libraries, digital publishing, educational services, and high quality software tools for the use of educational institutions around the world.
The Getty boasts highly competent software and infrastructure engineers, but the teams were having difficulty getting new product features into production fast enough. The engineering team was in charge of modifying the software, while the operating team needed to ensure stability and security. Though both teams were quite talented, the required compliance and governance structures were making it difficult to quickly make high-value changes to Getty’s digital products and services. “We needed a partner that could establish a low-friction platform for us to quickly deploy our microservices while maintaining reliability and security,” said Petrus Williams, head of infrastructure. “We also wanted to become self-sufficient in these new tools and processes and to avoid long-term dependency on a partner. So, we needed a partner to transfer their knowledge rather than build something we didn’t know how to operate. That’s when we discovered Blue Sentry.”
Blue Sentry embedded a fractional platform team to work alongside Getty engineers to enhance knowledge transfer and collaboration. A roadmap and Agile cadence were established so the teams could reach the end state more quickly. As part of the effort, Blue Sentry assisted The Getty team with pulling all of its infrastructure elements into Terraform modules so all future changes were approached with an Infrastructure as Code (IaC) process. From there, the transformational work began including:
- Building all-new EKS clusters in Terraform and training staff on adding new clusters and workspaces as needed
- Setup of their Terraform instance on Terraform Cloud for central management/auditing of all infrastructure changes
- Working with the developers to modify their current Docker files and applications to have configuration removed and to help build more 12-factor compliant artifacts. This was required to help support an immutable artifact setup
- Configuring ArgoCD for GitOps management and setup of continuous deployment to all of the EKS clusters
- Working with the developers to create a CircleCI re-usable orb to centralize the code needed to support GitOps and deploy new application versions to the cluster
- Moving 10 of their workloads to EKS. As Blue Sentry was getting The Getty team up to speed, six of these were a joint effort, with The Getty taking increasing responsibility with each new deployment. The final four workload deployments were handled completely by Getty with Blue Sentry observing. In this way, Getty demonstrated the skills acquisition and repeatability of all the work done for the initial applications
The IaC approach that has been deployed allows The Getty engineers to quickly make infrastructure changes while ensuring security and repeatability across environments. Maintaining this discipline will also greatly reduce the burden of audit and compliance. The 12-factor methodology and integrated gitflow tools allow Getty to quickly test and deploy new features without compromising reliability or security.
“We have gone from months to minutes in the time it takes to make a successful change,” said David Newbury, head of software. “But the best thing is we are now self-sufficient and fully capable of migrating our remaining portfolio. We could not have asked for a better partner to get us started.”