> ## Documentation Index
> Fetch the complete documentation index at: https://docs.incredibuild.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Utilization Dashboard

**Internal Article:** Currently this feature is only available to a few select customers upon CS approval.

The Utilization Dashboard provides valuable visual insights into your builds such as CPU time saved, daily active Initiators, and more. You can access the Dashboard by going to the Coordinator Manager and clicking the Dashboard icon.

This can help you discover areas you might need to investigate such as: 

* Determine if you are getting the maximum possible acceleration for your existing grid, or if additional resources could increase acceleration

* Verify that you have a large enough helper pool to meet your peak hourly demand

* Ensure that your Build Cache is being utilized optimally

* Learn about how much time is saved by Build Cache, and see if you can save even more time by purchasing additional Cache Cores

The data contains a summary at the top and a number of different graphs. Both the overview and the graphs share a single filter which you can use to control the time frame, or view data for a specific Initiator type, Build Group, or Build type.

You can view more information by hovering over each graph:

## Time Zones

When viewing the dashboards, the times displayed are from the perspective of the viewer's browser. This means that the times will look different depending on who is viewing the data.

## Utilization Dashboards

### Overview

The overview bar shows you the total number of builds that were run and their total duration over the selected time period.

### Helper Core Assignments and Requests

This graph shows you the maximum number of helper cores **assigned** to assist with running a build, measured every 15 minutes, compared with the number of **requested** helper cores. The **available** helper cores is the total available pool of potential helper cores and includes both assigned and unassigned helpers.

Each Build Group is closed environment in terms of helper availability, so we recommend using the Build Groups filter to view this graph for each individual Build Group.

* If available helpers are lower than requested helpers, you should increase your helper pool to improve acceleration.

* If assigned helpers are lower than requested helpers, but there were enough helpers available this may be because:

  * **Licenses:** Not enough licenses for available floating helpers

  * **Build Group Configuration:** helpers are available in Build Group A, but requested in Build Group B.

  * **Not enough helper machines:** If a helper is allocated to a assist a build, it cannot assist other builds at the same time (even if cores are available). Enterprise customers can use the [Multiple Helpers](/windows/enterprise-helper-multi-builds) feature to resolve this.

### Build Cache - Cache Core Usage

This graph shows cache cores usage over time, broken down by hour. To learn about how cache cores licenses work, see [License Model](/windows/build-cache-overview#License).

Cache core usage should correlate closely with the overall number of builds being run. You can use the [Build Dashboard](/windows/dashboard-builds)'s Total Builds and Build Run Time graphs to make this comparison. If these graphs do not correlate, you should investigate to make sure you are getting the most our of Build Cache.

### CPU Run Time and CPU Saved Time by Build Cache

Displays the total CPU time per day as well as the CPU time that was saved by using Build Cache.

### Daily Active Initiator Machines

Displays the number of initiator machines that ran at least one build in a particular day. If you see any machines that are missing licenses you should assign them licenses. Machines without licenses can run builds, but only locally without distribution or caching.

### Daily Active Helper Machines

Displays the number of Helper machines involved in builds each day. When using Incredibuild Cloud, the data is broken down by on-prem (physical) machines, and Cloud machines.