A policy allows you to modify the way that vRealize Operations analyzes and presents data in the dashboard and in various views and reports. A policy allows you to adapt the appearance and operation of vRealize Operations to your environment and preferences.

A policy includes the following settings:

• Metrics or properties for different object types (such as virtual machine, host, or Datastore)
• Symptom and alert definitions
• Badge threshold definitions
• Buffers, such as provisioning time buffers, and VMware vSphere® High Availability buffers

A comprehensive set of content is readily available when you install and configure vRealize Operations. You might have to customize some of the content to satisfy certain business requirements or to adhere to your company’s policies.

For example, you can create custom policies to support service-level agreements and business priorities. vRealize Operations allows you to modify existing content and create content if necessary.

You can customize the following content:

• Policies
• Dashboards
• Dashboard views
• Reports
• Alert Definitions
• Symptom Definitions
• Recommendations
• Super Metrics

Policies can be found by navigating to the Administration tab and then selecting Policies. Here you will find two tabs: Active Policies and Policy Library. Policy Library is a repository of standard policies delivered with vRealize Operations, policies installed with management packs, and policies created by a user or the installation wizard.

Let's take as an example vSphere Solution's Default Policy and look closer at its settings.

On the Policy Library tab, select vSphere Solution's Default Policy and click Edit. The Edit Monitoring Policy window opens: 

In the left-hand side pane, you can find the steps you must go through when creating a policy.

On the Select Base Policy page, select a base policy from which your new custom policy can inherit settings:

• The Start with drop-down menu allows you to choose which policy is to be used as a starting point. All settings from the starting point policy are inherited as default settings in your new policy.
• The Policy Preview pane displays tabs for the inherited policy configuration, and the configuration for the policy that you are creating. From this pane, you can view the number of enabled and disabled alert definitions, symptom definitions, attributes (metrics and properties), and the number of local changes. A local change is a change made to a setting in the policy that you are building which overrides the same setting in the parent policy.
• You typically apply a default policy to most of the objects in your environment.

Create a separate policy for each object group and make only minor changes in the settings for that policy so that vRealize Operations can monitor and analyze dedicated groups of objects:

On the Analysis Settings page, you can override settings for badge calculations. You need to click the padlock icon to the right before you can edit. In this example, I'm editing the Workload badge calculation. If you want to change the settings for an object defined in a management pack, you need to click the green plus sign next to Add settings for new set of objects and choose it from the drop-down list.

On the Workload Automation page, you can edit the Consolidate Workloads, Balance Workloads, and Cluster Headroom workload policies. Here is also where you can select whenever to prioritize virtual machines with the lowest or highest demand during balancing and whether to allow Storage vMotion or not.

The Override settings from additional policies drop-down menu enable you to select policy templates that you can add into the overall policy.

Built-in policies are provided for you to configure settings such as overcommit ratios, buffers, alerts, time periods, and peak usage periods. Choose from the built-in policies, policies provided by installed management packs, or policies that you created:

On the Collect Metrics and Properties page, you can find all the metrics and properties available in vRealize Operations. In your policy, you can enable or disable the collection of attribute types, which include metrics, properties, and super metrics.

These attribute settings are used by symptom definitions, specifically the metric-based and property-based symptom definitions. Since these settings affect the alerts that are generated and the results that are reported in dashboard scores, you must be aware of which symptoms are available that use the metrics and properties that you are customizing.

You can identify whether a metric, property, or super metric attribute is considered to be a Key Performance Indicator (KPI) when vRealize Operations reports the collected data in the dashboards. For example, you might identify disk command latency to be a KPI on a datastore cluster object. KPIs can be inherited from the starting point policy:

On the Alert/Symptom Definitions page, you can suppress a smart alert from an adapter that is not useful to you. You do this by changing the state of the adapter to Local. For example, out-of-the-box alerts are disabled for the Host is in maintenance for at least 72 hours alert definition.

Here is where you can also change a symptom definition threshold. You can do this by selecting Override from the Condition column.

For example, you can override at least one cluster in the custom data center has high CPU demand workload symptom definition and change the default threshold value that will generate an warning alert.

On the Custom Profiles tab, you can specify setup profiles of typical objects and the resources they consume, then vRealize Operations can show you how many of those objects will fit in your remaining capacity.

On the Apply Policy to Groups tab, you can select the groups to which to apply the policies.

As you can see, policies are an indispensable tool that can help you fine-tune what is being monitored in your environment and at what threshold value alerts should be generated for each and every metric in vRealize Operations.