Container Cluster Management landing page

Container Cluster Management enables observability of the health and performance of such containerized systems so you can optimize in accordance with your organization's needs.

The Overview page provides a canvas that you can populate with widgets from a library to create a personalized dashboard for monitoring your container clusters, tailored to the specific needs of your organization.

A new dashboard is available, displaying Utilization Insights, complementing observable data on this page and providing a distilled view limited to CPU and memory over-utilization and under-utilization.

Overview page

The landing page of the Overview page contains the following elements that enable efficient navigation and effective oversight of the Container Cluster Management:

Universal Header: The landing page header contains user-oriented information that assists with navigation.

Data and viewing selection: Filters and dashboard customization enable an at-a-glance understanding of the particulars, such as environments and providers of interest to you, as well as saved dashboard views that enable quick retrieval of information critical to your organization.

Actionable Insights and graphical data displays: Displays that provide actionable and otherwise meaningful data to enable quick and effective decisions regarding your managed clusters.

Container Cluster Management details table: A list of clusters with their associated details in tabular form.

Header

The dashboard contains multiple tabs that present distinct views tailored to individual requirements within the organization. Supported views include the following:

Capacity & Utilization Overview: Limited to subscribers assigned the Capacity & Performance Engineer role, this view enables capacity engineers to view clusters in the context of over- and under-utilization and overcommitted clusters. For more information, see Capacity Planning DashboardCapacity Planning Dashboard.

Container Cluster Management tab: Contains all landing page content described in subsequent sections of this topic.

APIVerse tab: Contains the Kyndryl Bridge - Container Cluster Management Controller Service APIs.

Global filter

Container Cluster Management supports a data filter to enable the selection of data you want to display in the dashboard, Actionable Insights and the Cluster Overview page. To filter displayed data, select the filter icon under the header. All graphic displays are adjusted to accommodate the filtering criteria.

The following filter categories are supported:

Providers: Implemented provider integrations.

Connections: Individual, created connections.

Applications: Individual, deployed applications.

Environments: All implement environments, such as development and staging.

Container type:

Kubernetes

OpenShift

By default, all data is displayed, and an All selection can be applied to any of the filter categories. Select one or more items from each filter category using the drop-down menu and click Apply to apply your filter selections or click Cancel to abort your selections.

When you apply your filter selections, the service displays the selected items in the filter field below the filter icon. You can dismiss any or all of the filter categories by clicking the X to the right of the category name; the service updates the dashboard accordingly.

Note: The global filter is so-called because it is present on all Container Cluster Management throughout the service.

User preferences dashboard customization

The dashboard view can be customized and saved, making it easy to compare applications' performance. Also, it is a great option to maintain the current statistic parameters and revisit them later.

The Clear view button serves as a reset button for the current customized filtered view; any unsaved customized view of the current dashboard will be returned to its default state when clear view is selected.

Note: The clear view option will be disabled when the default dashboard view shows all data; only when customizing a view from the dashboard will this option be enabled.

The Custom view button comes with an expanded selection of options that will allow more actions to be performed in the dashboard:

Save new view: By selecting Save new view, all current filters available will be saved, including the chosen timeframes and table settings selections. A confirmation window will require inputting a name for the customized dashboard view and selecting whether you want the current default view.

Update view: selecting this option allows the update of the current dashboard, including the filters, timeframes, and table settings using different parameters; this is not applicable if the current view is from a default dashboard.

Delete view: this option will delete the current customized dashboard view.

Note: If all customized filters are deleted, the Container Cluster Management displays the default stage dashboard view. This is the view before customization. Also, If an user customizes the dashboard filters but the view is not saved or updated before the end of the session, the system will not honor these customization changes upon the next login session.

Graphic displays

Container Cluster Management landing page presents critical cluster information at a glance in the form of four widgets and a table view:

Alerts

On the Dashboard page, the Alerts section provides all the alerts for all the clusters controlled by the user across the various cloud platforms. The user can navigate to the cluster alerts page to evaluate the cluster alerts from any available cluster table view by selecting the desired cluster on the table.

Operational Insights

The landing page contains four actionable insights widgets that provide insights into your IT estate:

Security and Compliance

Performance

Reliability

Optimization

Each widget navigates to the Insights Summary by category page. The page provides statistical data to help you assess the state of your IT estate using status data and an access point to details regarding specific clusters, complete with configuration recommendations. Using it, you can balance resources and take preemptive actions as needed, enabling a more nimbler approach to reliability and efficiency.

More information is available at Actionable InsightsActionable Insights.

Cluster health by provider

The donut chart shows information about the health of the clusters for various providers. The size of the donut chart will be in ratio to healthy and unhealthy clusters. Healthy clusters will be represented in green color, Unhealthy clusters will be shown in red.

By hovering over a cloud provider on the Heatmap, the tooltip provides:

The total number of clusters.

The number of healthy and unhealthy clusters.

The cloud provider's name. When you choose a provider, the donut chart displays all of the clusters for that provider. The tooltip displays the full cluster name, the total number of clusters and the percentage of healthy and unhealthy clusters.

Clusters health by region

The Geomap widget displays information about cluster health based on their location. On the Geomap, the number of clusters in each region is displayed. When you hover your mouse over a clustered region, the tooltip reveals the region's name and the number of healthy and unhealthy clusters in that location.

Interaction between Geomap and Heatmap

When a specific provider is selected in the Cluster health by provider widget, the cluster table will be updated with the information about the cluster or clusters within your selection. The same behavior will occur when the provider filters the Cluster by region widget.

Clusters table

The Running Clusters table displays the following information about the cluster:

Cluster Name: The complete cluster name is displayed in this column.

Applications: Applications dependent on that cluster.

Environment: The name of the environment hosting the cluster.

Connections: The name of the running connection for the cluster.

Provider: The name of the cloud provider is displayed in this column.

Location: Displays the code of the Location of the cluster.

Health: Presents either Healthy or Unhealthy designation dependent on predefined health criteria. Refer to the subsequent Health Column section.

Alerts: Displays the number of critical alerts for a cluster.

Version: Displays the version of Kubernetes used by the cluster.

All the columns in this table can be sorted except the Connections and Location columns. Above this table, you will find a search box that allows searching for clusters by name and a Settings icon that allows changing the table settings to show or hide the additional columns (CPU Usage, Memory Usage and File Usage).

Health Column

The Health column contains a binary designator of each cluster as Healthy (green) or Unhealthy (red). If a cluster is healthy, the designator is inactive. If a cluster is deemed unhealthy, the designator is active, allowing users to click the designator and navigate to a details page that provides specific information on which health thresholds have been breached.

The following criteria are used to gauge cluster health:

Node Health: Displays how many nodes are deemed unhealthy. The default threshold is 50%.

Critical alerts: Displays how many critical alerts have been issued. The default threshold is 1%.

Memory overcommitted: Displays the amount of memory that could be consumed by individual containers as a percentage of the set memory limit for that cluster. The default threshold is 90%.

CPU overcommitted: Displays the amount of CPU that could be consumed by individual containers as a percentage of the set CPU limit for that cluster. The default threshold is 90%.

All health thresholds are set as an administrative function. Refer to Setting cluster health parametersSetting cluster health parameters for details.

Container Cluster Management details table

The Table View supports detailed views for each cluster. To access details for a specific cluster, click the cluster name for that cluster. The service navigates to the details page for the selected cluster. The left navigation bar presents multiple detail selections as follows:

Cluster: Tab presenting data about each Cluster. For more information, see Cluster.

Actions: Enables taking an action by way of the Kyndryl actions service. For more information, see Cluster details pageCluster details page.

Workloads: Tab presenting applications running in Kubernetes. For more information, see Workloads.

Network: Displays available Services and Ingress objects. For more information, see Network.

Configuration: Shows details about configmaps and secrets. For more information, see Configuration.

Storage: Shows the persistent volumes of the clusters. For more information, see Storage.

Access Control: Shows the Access Control details of the cluster. Cluster access control.

Policy: A cluster policy restricts the ability to configure clusters based on rules that limit the attributes.

Custom Resources: Custom resources are Kubernetes API extensions that allow objects to be created, updated and deleted. For more information, see Custom resources.

Capacity Planning Dashboard: Gain a keen understanding of current capacity and the ability to preemptively balance resources for maximum cost efficiency.

Executive dashboard: View a summary of critical performance data impacting applications to assist in removing obstacles.

Cluster details page: Learn about this resource for a deep view into individual clusters, to acquire a keen understanding of performance and health.

Setting cluster health parameters: Learn how to set parameter thresholds as the baseline for cluster health.

Integration with applications Container Cluster Management: Find out how to integrate your application deployments with Container Cluster Management to observe status, health and regional concentrations of your container clusters.

Supported hybrid IT estate personnel: Understand the target users Container Cluster Management supports.

User and role management: Learn how to grant appropriate access to your users in accordance with the type of cluster data they need.

My Dashboard: Learn how to customize your personal dashboard to tailor disparate data you most commonly need to a common view.

Missing data notifications: Understand the status of missing data through automated notifications.

Glossary terms: Understand the parlance of Container Cluster Management to comprehend the nuances and details of container clusters managed by users of Container Cluster Management .