Cluster Overview

Cluster Overview is your home page for the Sosivio Dashboard. Here you get a high level overview of what is happening in your cluster. This includes all active recommendations from the Sosivio Prediction Engine as well as health scores for your cluster including specific scores for the most important parts of your environment - the Platform, Applications, and Deployment.

Use the AI Predicted Failures Graph to observe as critical events matures in severity and take action before they turn into failures. From here you can also dive deeper into any entity, object, or component which requires your attention.

The Node Load Average widget shows the nodes with the highest load average for the past 5 minutes. This enables an easy way to see which noads are working the most and if any action is needed. The High-Priority Resource Reductions (Over-Allocated) widget displays the applications wasting the most resources. This enables user to see the highest amount of over-allocated applications and easily reduce them to free up space on your cluster. The High-Priority Resource Additions (Under-Allocated) widget displays the applications in need of additonal resources. This enables users to see the highest amount of under-allocated applications and easily increase them to instantly improve performance and stabiltiy of the throttling application.

Let Sosivio be your guide - use the different dashboards for recommendations and in-depth analysis of your cluster.

AI Predicted Failures Graph

Here you can see all of the Sosivio Failure Predictions in real time. Sosivio shows events to you before they materialize into failures so that you can prevent them.

Each circle on the graph represents a potential failure event. Events progress along the graph as the Confidence and Severity levels of Sosivio's prediction increases. Each sequence is color coded according to the severity of the event.

A yellow event is a warning
An orange event is an urgent warning
A red event is a failure

You can click on the graph to go to the command center to see full detailed warnings and recommendations.

Cluster Pods, Nodes, and Namespaces

Sosivio also displays information on the current number of Pods, Nodes, and Namespaces.

Pods

Pods are displayed as:

Running
Down
Inactive

Nodes

Nodes are displayed as:

Running
Down
Unscheduled - Nodes that will not be used to deploy new Pods on them

Namespaces

The current number of namespaces on your Kubernetes cluster.

Health Scores

The health scores are displayed in three separate categories:

The Platform
The Applications
The Deployment

The Platform is your Kubernetes cluster itself, the Applications are the processes in your Pods, and the Deployment covers the application architecture and configuration.

On the bottom portion of the Cluster Overview page are the Node Load Average, High-Priority Resource Reductions (Over-Allocated), and High-Priority Resource Additions (Under-Allocated) widgets. The Node Load Average widget shows the nodes with the highest load average for the past 5 minutes. This enables an easy way to see which noads are working the most and if any action is needed. The High-Priority Resource Reductions (Over-Allocated) widget displays the applications wasting the most resources. This enables user to see the highest amount of over-allocated applications and easily reduce them to free up space on your cluster. The High-Priority Resource Additions (Under-Allocated) widget displays the applications in need of additonal resources. This enables users to see the highest amount of under-allocated applications and easily increase them to instantly improve performance and stabiltiy of the throttling application.