Understanding the effect of optimizer thresholds on views, models, and optimizer rules

With the Optimizer metric thresholds, you can monitor the behavior of a metric with respect to a specified threshold value. You can define Good and Warning thresholds for a metric. For details, see Configuring-and-managing-thresholds-for-metrics-and-indicators. 

When you add or modify the thresholds, it affects various functionalities of BMC Helix Continuous Optimization. Some functionalities use only the warning thresholds, while other functionalities use both thresholds. Use the information in this section to understand how Optimizer metrics thresholds are used in these functionalities. 

Views

The Optimizer thresholds affect the following functionalities in the Views tab.  

Out-of-the-box views 

In Out-of-the-box views, the values set in the Optimizer thresholds are used by the bar renderer of all utilization metrics. The color of each bar renderer corresponds to a utilization status according to the good and warning threshold values that are set in the Optimizer: green for a value less than a good threshold, yellow for a value between the good and warning threshold, and red for a value that crossed the warning threshold.

These renderers are displayed in the entity summary table or details pages.

Example summary page for vSphere Cluster:

Example details page for vSphere Cluster:

The vertical bars in the renderers of the details page indicate the good and warning threshold values that are set in the Optimizer. 

The renderers in the entity summary and details pages of the following out-of-the-box views are affected by optimizer thresholds: 

• 
  
  • Capacity-Pools-view
  • AWS-views
  • Azure-views
  • GCP-views
  • IBM-views
  • vCloud-view
  • Pivotal-Cloud-Foundry-view
  • Kubernetes-view
  • OEM-view
  • Server-views
  • Storage-views
  • AIX-views
  • KVM-views
  • vSphere-views

Business Services view

The Business Services view uses Risk, Usage, and Days to saturation indicators for each service pool that you have defined. These indicators are based on the Optimizer threshold values. You can change these thresholds from the Business Service view settings page. For details, see Configuring-the-business-services-view-settings.

The color of the bubble chart, Risk and Usage charts, and each bar renderer in the details page correspond to the good and warning threshold values that are set in the Optimizer. 

In the bubble chart, the saturation of business services is based on the saturation of the associated service pools. The Optimizer threshold values configured for these service pools or associated servers are used for indicating whether the service pool and, therefore the business service is saturated.

The color of each bubble corresponds to the saturation status of a business service. 

• When long term forecasted saturation is not used: 
  • Already saturated: Red
  • No saturation up to 90 days: Green
  • Every thirty days in-between: different shades of yellow and orange
• When long term forecasted saturation is used: 
  • Already saturated: Red
  • No saturation up to 180 days: Green
  • Every thirty days in-between: different shades of yellow and orange

Similarly, the Risk and Usage scores for business services are based on the Risk and Usage score of the associated service pools.

For details on the Risk, Usage, and Days to saturation indicators used in the Business Service view, see Indicators-used-in-the-Business-Services-view.

Example Bubble chart in the Business Service Overview page

Example Usage and Risk values for pools

Custom views and Summary data marts

The Optimizer thresholds can be used in a custom view through a summary data mart. 

1. In the Advanced editor of the Summary Data Mart wizard, set the includeThresholdColumn as true.
   The Optimizer thresholds set for the selected entity type (for example, Virtual Cluster - VMware) get included in the summary data mart.
2. Use this data mart while creating a custom view of the Summary table or the System details template.
3. Configure a table column or Summary ribbon for the utilization metric (for example, CPU_Utilization).
4. Edit the parameters and use good and warn thresholds columns for the bar renderer.
   The Optimizer thresholds get included in the custom view. 

The colors of the bars indicate whether the values are below, between, or above the Optimizer thresholds:

• Green: Indicates that the value is below good thresholds.
• Yellow: Indicates that the value is between good and warning thresholds.
• Red: Indicates that the value is above the warning thresholds.

Example bar renderer configuration for a CPU_Utilization metric 

Out-of-the-box recommendations

The Future saturation recommendation detects entities (for example, virtual clusters and hosts) for which the bottleneck resource usage has already exceeded thresholds or is expected to exceed them. All future saturation recommendations (VMware or Kubernetes) use time forecasting models. The warning Optimizer threshold value configured in time forecasting models for these metrics is used for detecting saturation and recommendation. The recommendation severity and color are based on the Optimizer rules and can be configured. For example, the severity of the bottleneck resource and recommendation is marked as critical (red) if the resource saturation is expected to occur within the next 7 days or warning (yellow) if it is expected to occur within the next 8 to 45 days.

For details, see Future-saturation-recommendation and Kubernetes-Forecasted-saturation-recommendation.

Time forecasting models

When you create a time forecasting model, by default, the specified optimizer thresholds for that metric are applied to the model. Based on the thresholds, the model shows the prediction data of the selected metrics.

For details, see Creating-a-time-forecasting-model.

Custom optimizer rules

While creating an optimizer rule based on custom conditions, the warning optimizer threshold can be used in the following scenarios: 

• Data vs Threshold condition type when Input data is based on historical data: When the optimizer thresholds option is selected, this condition compares a metric of your system with a warning threshold set in the optimizer. For example, you can configure a rule that generates an alert if the average value of samples at summary resolution in the period Feb-Mar 24 is greater than 70.
• When input data is based on data marts: When the optimizer thresholds option is selected, this condition compares a metric of your system with a warning threshold set in the optimizer.

For details, see Adding-a-custom-Optimizer-rule.