Case study · Fintech · Platform migration
Automated deployments with an EKS migration for Klub, one of India's largest revenue-based financing investors.
Klub · Revenue-based financing
ECS limited how far they could scale and locked them to one provider. We moved the platform to EKS to remove both constraints.
The opportunity
A platform that could not scale past its own orchestrator.
Klub provides growth financing to high-affinity brands, using financial innovation, community engagement and deep data-driven analytics to put skin-in-the-game capital behind entrepreneurs. Its applications ran on Amazon ECS.
That became the constraint. ECS limited scalability for large-scale applications and traffic spikes, it was proprietary enough to lock Klub into a single cloud provider, and the applications' own architecture carried a lot of platform dependencies on top.
Limited scalability
ECS made large-scale applications and traffic spikes difficult to handle.
Vendor lock-in
A proprietary orchestrator meant migrating providers was not straightforward.
Complex application architecture
The applications carried a lot of platform dependencies.
The impact
Scale, portability and automated deployment, together.
Greater scalability
EKS lets Klub manage critical financial components and scale its event-driven architecture independently.
Cloud-agnostic by design
Kubernetes means moving to another provider is possible rather than a rebuild.
Deployment automated
The time and effort required to manage applications dropped once deployment was no longer manual.
Scaling that responds
Horizontal Pod Autoscaler and the Kubernetes Dashboard monitor and scale applications as needed.
The stack
What it runs on.
- Orchestration
- Amazon EKS, Kubernetes
- Infrastructure as code
- Terraform
- Traffic
- Ingress controller
- Scaling and visibility
- Horizontal Pod Autoscaler, Kubernetes Dashboard
The road ahead
The platform is portable. Next is what runs on it.
With containers on Kubernetes and deployment automated, the delivery pipeline and cost work that followed were additions to this foundation rather than separate migrations.
- 01
GitOps delivery on Argo CD
- 02
Cost optimisation across the estate
- 03
Independent scaling per service
Talk to us
Bring one process. Leave with an outcome architecture.
30 minutes on one function, mapped to the architecture behind it. No slides.