Case Study: Capital One achieves faster, more resilient decisioning with Kubernetes from CNCF

Capital One Supporting fast decisioning applications with Kubernetes

Capital One, a major US retail bank, sought to build a resilient and high-speed provisioning platform on AWS for its critical applications involving big-data decisioning, fraud detection, and machine learning. Their challenge was to support millions of daily transactions while ensuring rapid deployment and the ability to completely rebuild their clusters from base AMIs for security.

The Cloud Native Computing Foundation's Kubernetes provided the strategic solution, used as a foundational substrate for the platform. This implementation by the Cloud Native Computing Foundation resulted in massive improvements: time to market for a new application decreased from a quarter to a fortnight, deployment frequency increased by orders of magnitude, and the cluster rehydration process was reduced from a day to a couple of hours. Furthermore, Capital One estimates that without Kubernetes, their pure AWS expenses would be three to four times higher.

Capital One

Keith Gasser

Lead Software Engineer