Service Insights

Download this manual as a PDF file

The Service Insights page (Business Services > Service Insights) provides a comprehensive dashboard for monitoring the health, usage, and performance of all business services in your organization. This page aggregates key metrics and visualizations that help you quickly assess service adoption, identify performance bottlenecks, and troubleshoot issues affecting service hierarchies and calculations.

Rather than navigating through individual services to gather operational insights, the Service Insights page displays aggregate statistics and performance trends across your entire service portfolio. You can use this page to identify the services consuming the most resources during hierarchy and Health, Availability, and Risk (HAR) calculations, spot services with missing constituents, and monitor the time efficiency of your service management engine.

Use the following menu options to navigate the Skylar One user interface:

  • To view a pop-out list of menu options, click the menu icon ().
  • To view a page containing all of the menu options, click the Advanced menu icon ().

Prerequisites

To use the Service Insights page (Business Services > Service Insights), ensure that your Skylar One deployment meets the following requirements:

  • Skylar One version 12.5.20 or later

  • The BIZ_SVC_INSIGHTS_VIEW access hook assigned to your user account

  • The ITS_SERVICE_VIEW access hook assigned to your user account

Viewing the Service Insights Page

The Service Insights page (Business Services > Service Insights) is organized into several sections, each providing a different view of your service environment's health and performance. The page layout includes an overview information bar at the top, followed by multiple widgets organized in a grid below. All metrics and charts on the Service Insights page include tooltip information. Hover over any chart element or metric to view additional context and details about the data displayed.

Service Overview Information Bar

The information bar at the top of the Service Insights page displays an overview of all services in your organization. The following metrics are included in individual cards:

  • Total Services. The total number of all services in your organization

  • Root Services. The number of top-level services with no parent

  • Aggregate Services. The number of services that aggregate other services

  • Device Services. The number of services that monitor devices

  • Shared Services. The number of services shared across multiple organizations

  • Disabled Services. The number of services that are currently disabled

  • Empty Services. The number of services with zero constituents

  • Orphaned Services. The number of services with no parent service

Service Insights Widgets

The Service Insights page (Business Services > Service Insights) includes the following widgets, displayed in a grid layout below the overview information bar. Widgets are organized to group related metrics together, allowing you to quickly identify performance issues and service characteristics across your entire service portfolio.

All widgets on the Service Insights page include tooltip information. Hover over any chart element, data point, metric label, or icon to view additional context and details about the data displayed.

Top 10 Services by Constituent Count

A horizontal bar chart displaying the top 10 services with the highest number of constituents. This widget helps you understand the complexity and scope of your service hierarchies.

What this shows:

  • The number of child services or devices included in each service

  • Services with the largest footprint in your service model

  • Candidates for potential optimization if constituent counts approach the 100-constituent limit

How to interpret:

  • Services with high constituent counts consume more resources during HAR calculations. If a service is approaching or exceeds 100 constituents, it may require redesign or subdivision into multiple smaller services. The color coding on the bars indicates the current health or severity status of each service.

Related metric:

  • Compare this data with the Top 10 Services by HAR Calculation Time widget to identify which complex services are most resource-intensive.

Top 10 Services by HAR Calculation Time

A horizontal bar chart displaying the top 10 services with the longest Health, Availability, and Risk (HAR) calculation times. This widget directly reflects the performance impact of individual services on your system.

What this shows:

  • The amount of processing time (in seconds or milliseconds) required to calculate HAR values for each service

  • Services that consume the most resources during HAR refresh cycles

  • Performance bottlenecks in your service management engine

How to interpret:

  • Services appearing here are the top consumers of the Business Services: Service Management Engine processing time. If these services have high constituent counts or complex service policies with many rules, they are likely responsible for long calculation times. Optimizing these services, by simplifying policies, reducing constituents, or adjusting the refresh interval, can improve overall system performance.

Recommendations:

  • Review service policies for services in this list to identify unnecessarily complex rules

  • Consider subdividing services with high constituent counts

  • Adjust the refresh interval for services that do not require frequent updates

Top 10 Status Policies by Metric Rules

A horizontal bar chart highlighting the top 10 status policies based on the number of metric rules they contain. This widget helps you identify policy complexity across your service environment.

What this shows:

  • The number of health, availability, and risk rules defined in each policy

  • Which policies are most complex

  • Policies that may benefit from simplification or review

How to interpret:

  • Policies with many rules are more computationally intensive and may contribute to longer HAR calculation times when applied to multiple services. High rule counts do not inherently indicate a problem, but they are worth reviewing to ensure each rule provides necessary value. Duplicate rules, overly granular conditions, or rules that rarely trigger may be candidates for removal.

Related metric:

  • If policies from this list are assigned to services appearing in the Top 10 Services by HAR Calculation Time widget, policy simplification could improve performance.

Services with Zero Constituents

A searchable widget displaying all services that have zero constituents (empty services). You can search for services by name using the search filter field at the top of the widget.

What this shows:

  • A list of all services that contain no child services or devices

  • Services that do not contribute to HAR calculations

  • Services that may be incomplete, accidental, or no longer needed

How to interpret:

  • Services with zero constituents do not generate HAR values and do not impact monitoring. They typically represent one of the following:

    • Newly created services awaiting constituent assignment

    • Placeholder services created during service model design but not yet populated

    • Orphaned services whose constituents were removed or deleted

    • Archived or unused services that should be deleted

Recommended action:

  • Review services in this list and determine whether they should be populated with constituents, assigned to a service policy, or deleted. Services with zero constituents can also indicate incomplete service design or accidental removals that may require investigation.

Time Taken to Complete Topology Calculation

A line graph showing the time (in seconds) taken to complete hierarchy calculations over a period of time. The x-axis represents time in hourly intervals, and the y-axis represents elapsed time in seconds. This metric reflects the performance of the Business Services: Service Topology Engine process.

What this shows:

  • The duration of the Business Services: Service Topology Engine process, which runs every 5 minutes by default

  • Performance trends in calculating service relationships and hierarchies

  • Periods of increased load or complexity in your service model

How to interpret:

  • Consistent low values: Your service model relationships are stable and efficiently calculated. The topology engine is performing well.

  • Occasional spikes: Temporary increases in calculation time may occur after bulk service changes, configuration updates, or device additions. These spikes are normal if they return to baseline afterward.

  • Sustained high values: Prolonged elevated calculation times may indicate that your service topology is becoming too complex, that you have exceeded recommended limits (such as 2,500 services), or that your system resources are constrained. Consider optimizing your service model or increasing system resources.

Related metric:

  • Compare this with the Time Taken to Complete HAR Refresh widget. If topology calculation time is high, HAR calculation time may also be affected.

Time Taken to Complete HAR Refresh

A line graph showing the time (in seconds) taken to complete Health, Availability, and Risk (HAR) refresh operations over a period of time. The x-axis represents time in hourly intervals, and the y-axis represents elapsed time in seconds. This metric reflects the performance of the Business Services: Service Management Engine process, which aggregates metric information to compute HAR values.

What this shows:

  • The duration of the Business Services: Service Management Engine process, which runs every 15 minutes by default

  • Performance trends in calculating health, availability, and risk for all services

  • Overall system efficiency in processing service metrics

How to interpret:

  • Consistent low values: Your HAR calculations are efficient, and the service management engine is performing well.

  • Gradual increase over time: As you add services or modify policies, calculation time may gradually increase. Monitor this trend; if it continues to rise, optimization may be needed before calculation times become problematic.

  • Sudden spikes: May indicate temporary system load, a large batch of service configuration changes, or collection of metrics from a high volume of devices.

  • Values exceeding 15 minutes: If HAR calculations take longer than the 15-minute refresh interval (the default), calculations will queue up and HAR values will become stale. This situation requires optimization.

Recommended monitoring:

  • Track this metric regularly. If values trend upward, proactively optimize services, policies, and refresh intervals to maintain performance.

Orphaned Services

A searchable widget displaying all orphaned services (services with no parent service). You can search for services by name using the search filter field at the top of the widget.

What this shows:

  • A list of all services that have no parent service assigned

  • Services that exist at no defined hierarchical level

  • Potentially incomplete or incorrectly configured services

How to interpret:

  • Unlike services with zero constituents, orphaned services may have constituents (child services or devices) but no parent. They represent a break in the expected service hierarchy. Orphaned services typically result from one of the following scenarios:

    • Accidental deletion: A parent service was deleted, leaving child services orphaned.

    • Incomplete service design: A service was created but not properly integrated into the service hierarchy.

    • Configuration error: A service was misconfigured and is not linked to an expected parent.

Recommended action:

  • Review orphaned services to determine the appropriate parent service. You can either assign a parent service through the service editor or delete the orphaned service if it is no longer needed. Orphaned services should be addressed to maintain a clean, understandable service model.