Hari C. Ayada

KRMOps (Kubernetes Resource Model Operations) is a Kubernetes-native approach to managing cloud resources declaratively. It extends the Kubernetes Resource Model (KRM) beyond cluster-native resources to AWS services and infrastructure, enabling fully API-driven cloud provisioning and reconciliation.
KRMOps is powered by ACK (AWS Controllers for Kubernetes) and kro (Kube Resource Orchestrator), both are now available as EKS Capabilities.
What are EKS Capabilities?
EKS Capabilities are fully managed, Kubernetes-native features that run within Amazon EKS, eliminating the need to install, maintain, and scale critical platform components on your worker nodes. This workshop focuses on ACK and kro capabilities:

• ACK - Enables management of AWS resources (S3, RDS, IAM, etc.) using Kubernetes APIs with continuous reconciliation to correct drift.
• kro - Allows you to define custom Kubernetes APIs that compose multiple resources into higher-level abstractions using simple configuration and CEL (Common Expression Language) .

With EKS Capabilities, AWS handles all operational aspect: installation, updates, scaling, and patching, while these services run within EKS rather than consuming your cluster resources.
How KRMOps Works with EKS Capabilities
Traditional cloud provisioning relies on imperative approaches (e.g., AWS SDK and CLI). With KRMOps and EKS Capabilities, everything is defined using Kubernetes-style YAML manifests that are continuously reconciled. The combination of ACK and kro capabilities enables:

• Declarative AWS Resource Management (via ACK capability):
  • Define AWS services using Kubernetes-native objects.
  • Apply them using kubectl, just like Kubernetes-native resources.
  • Continuous reconciliation automatically corrects drift in AWS resources.
• Automatic Dependency Resolution (via kro capability):
  • kro ensures resource dependencies are met before provisioning.
  • It dynamically updates configurations based on state changes.
  • Compose both Kubernetes and AWS resources into unified abstractions (RGDs).
• Self-Healing Infrastructure:
  • ACK continuously reconciles AWS resources to match desired state.
  • kro ensures Resource Graph Definitions remain consistent.
  • Failures are monitored and corrected without manual intervention.
• Zero Operational Overhead:
  • Capabilities run within EKS, not on your worker nodes.
  • AWS manages updates, scaling, and security patching.

The goal of this workshop is to demonstrate the powerful features and benefits of using kro to define, provision, and manage AWS resources directly from your EKS clusters. This workshop will walk through real-world examples, showcasing an end-to-end workflow for orchestrating cloud resources using kro’s Resource Graph Definitions (RGDs). The entire workshop takes about 2 hours to complete.
What You Will Learn
The modules in this workshop are designed around the most frequent use cases for EKS Capabilities in Kubernetes-based AWS environments by Platform Engineers and Developers. You will gain hands-on experience in:

• Using ACK capability to manage AWS resources (S3, RDS, IAM) declaratively from Kubernetes.
• Defining and deploying Resource Graph Definitions (RGDs) with the kro capability to compose multiple resources.
• Deploying a sample web application that dynamically provisions and interacts with AWS-managed services.
• Implementing IAM-based authentication for applications running inside the EKS cluster.
• Using nested RGDs to simplify complex cloud resource orchestration.
• Understanding how fully managed capabilities eliminate operational overhead.

Why Use EKS Capabilities?
EKS Capabilities introduce a powerful declarative approach for managing AWS services within Kubernetes with zero operational overhead. With ACK and kro capabilities, you can:

• Manage AWS resources using native Kubernetes APIs with continuous reconciliation (ACK).
• Define cloud resources as modular and reusable components using RGDs (kro).
• Dynamically manage dependencies between AWS services and Kubernetes workloads.
• Eliminate operational burden - AWS handles installation, updates, scaling, and patching.
• Free up cluster resources - Capabilities run within EKS, not on your worker nodes.

By the end of this workshop, you will have hands-on experience with EKS Capabilities and understand how they simplify AWS resource management using a fully managed, Kubernetes-native approach.
Let’s get started!