Amazon Elastic Kubernetes Service (EKS) is a managed Kubernetes service that facilitates smooth running of Kubernetes on both AWS and on-premises. It helps guarantee high availability of your clusters and its resources across numerous availability zones. EKS clusters need to be monitored extensively to ensure hassle-free operations of various mission-critical applications running on them.
Applications Manager is a comprehensive EKS monitoring tool that provides in-depth insights about EKS clusters, namespaces, nodes, and pods. It identifies performance bottlenecks, evaluates your clusters' ability to scale up and launch new applications, and notifies you of errors on the go.
Become aware of Cluster Status and get to know basic cluster information such as Cluster ARN and Cluster Endpoint at a single glance. Keep tabs on cluster usage and get notified of increasing Cluster Memory/CPU usage. If this value keeps increasing, you may need to plan and implement scaling of cluster capacity as and when required.
Tracking nodes and their corresponding CPU, memory and disk usage is pivotal to finding and fixing issues at the application level. Track Memory Limit and Memory Request metrics to understand overall memory usage in the node and ensure that it does not exceed the Allocatable Memory. These details help you to identify oversubscribed nodes and nodes under memory pressure, both of which can affect the cluster's ability to run workloads effectively. Similarly, track node CPU usage constantly to avoid node throttling.
Furthermore, details about the pods in the nodes and important configuration details are also available at your disposal.
Monitor critical pod related performance metrics and become aware of its state. Get notified of increased memory usage in pods to ensure they don't get terminated. Isolate pods with spiking memory requests. They may indicate improper configuration.
Visualize top 10 pods based on memory and CPU in graphical format to get a glimpse of overall performance of the pods at a single glance.
You can also get insights into container related metrics. Find and fix issues on problematic containers by getting notified on the Container Restarts metric.
Track the services running in your clusters and become aware of metrics such as Service Type, Service Protocol, and Running pods in service. Deployment details and its availability are also available at your disposal.
Get exclusive persistent volume (PV) details such as the status of PV, claim, capacity and the storage class it belongs to. You can also track Persistent Volume Claim (PVC) details such as its status, the capacity, etc.
Use the Service Map feature to visualize all service and namespace details at a single glance. The graphical display helps you to discover various components of the services and their status. Identify namespaces that are available and unavailable by recognizing the color codes in which they are represented. Hover over the service to get details such as host IP, port and number of running pods.
With Applications Manager, you can gain deep visibility into the performance of Amazon Elastic Kubernetes Service. Get started now with a full-fledged, free, 30-day trial now!
On-premise Kubernetes is an open source container orchestration platform which facilitates the automation of multiple manual processes that involve deploying, managing, and scaling containerized applications.
Google Kubernetes Engine (GKE) is the first managed native Kubernetes service that was created for scaling, deploying and managing containerized applications in the cloud.
Amazon's Elastic Kubernetes Service (EKS) is a cloud based Kubernetes service that is fully compatible with native Kubernetes and facilitates smooth running of Kubernetes on both AWS and on-premises.
Unlike other popular EKS monitoring tools in the market, Applications Manager monitors AWS EKS APIs, container insights and Kubectl data helping you to get deep insights about your service and its performance.
Yes, there are some pre-requisites for adding an Amazon EKS monitor. You can read more about the same here.