Kubernetes to the Rescue - GumGum's Million-Dollar Performance Leap
summary
GumGum modernizes its AWS footprint and claims ~$1,000,000 annual savings by standardizing EKS clusters, moving from one monolithic account to a multi-account split, and replacing ad hoc spot tooling with Carpenter node pools and consistent Terraform/Terragrunt modules. Observability uses Amazon Managed Prometheus and Amazon Managed Grafana with centralized alerts; identity uses AWS Identity Center; drift detection and GitHub Actions rollouts enforce parity. Measured wins include isolating a misconfigured VPC that saves ~$180k/year and cutting Japan-to-Singapore data transfer from $13k to $4k monthly (~$100k/year).
AI-written summary · about
chapters (9)
- 0:00Introduction and goals
- 1:03Company overview and savings goal
- 2:07Past pain points with Kubernetes
- 9:34Modernization with multi-account setup
- 11:18EKS add-ons and observability
- 13:02Infrastructure as code and GitOps
- 14:51Key advantages of modernization
- 17:16Data transfer cost savings case
- 19:07Q&A session
related talks
33mSpace Invaders for Kubernetes: Consolidating 50 Clusters at Scale
53mImplementing Kubernetes on with confidence
14mBuilding a high-performance analytics ecosystem on Amazon EKS
29mThe Frugal k8s Engineer: Strategies for Amazon EKS
15mGoing Multi Cloud in a Hurry with Quality and Style
27mKubernetes Everywhere: Lessons Learned From Going Multi-Cloud
47mBuilding sustainable architectures with Kubernetes
01h 02mLearnings from the Enterprise-level Kubernetes transformation at Elkjøp Nordic