You observe that during peak hours, the resources used in Sunbird ED are not enough for the deployed microservices. What should you do?

Sunbird ED's microservices are deployed on OCI Container Engine for Kubernetes (OKE). When resources are insufficient during peak hours, it means the existing cluster nodes cannot handle the load. The Kubernetes Cluster Autoscaler is the specific feature designed to solve this by automatically adding or removing worker nodes in the node pool based on the resource demands of pending pods. This ensures the microservices have the necessary compute resources to function correctly during high-traffic periods without manual intervention.

A. Create Instance Pools for the virtual machines (VMs) that run the non-containerized workload: This addresses non-containerized workloads, while the question specifies microservices, which are containerized on OKE.
C. Enable Auto-Tiering of the Object Storage bucket: This is a storage cost-optimization feature and does not affect compute capacity for microservices.
D. Create Volume Groups for the VMs: This is a block storage management feature for backups and cloning, not for scaling compute resources.

1. Oracle Cloud Infrastructure Documentation. "Using the Cluster Autoscaler." Oracle Help Center, Accessed May 2024. (This document explains that "The Kubernetes Cluster Autoscaler automatically adjusts the size of a cluster's node pools based on the resource requests of pending pods.")

2. Oracle Architecture Center. "Deploy Sunbird on Oracle Cloud Infrastructure" (Reference Architecture ID: OCI-ARCH-SUNBIRD). oracle.com, Section: "Recommendations", Subsection: "Scalability". (This document recommends using OKE's autoscaling capabilities to handle varying loads for the Sunbird deployment.)