Insights is the analysis layer that turns collected environment data into clear, actionable recommendations. It reviews real usage, performance, and configuration data and highlights where changes can reduce cost, improve performance, or simplify the environment.

Insights does not make changes automatically. It provides evidence-based recommendations that operators can review and act on.

NOTE: Insights is a live data feature and will only be available to live workspaces. It will not be available if your workspace is frozen since once a workspace is frozen it will roll into the history. For more info on creating a new live workspace, please click here.

Upsize

Recommended when a resource consistently runs short of capacity.

Typical triggers include high CPU utilization or memory pressure.

Goal: restore performance and stability by allocating more capacity.

Downsize

Recommended when a resource is consistently underutilized.

Common indicators include low CPU usage or excess memory allocation.

The recommended size still leaves enough capacity for peak usage, so a resource is not reduced below what it needs during its busiest periods.

Goal: reduce cost without impacting performance.

Resize

Recommended when the balance of resources is incorrect.

For example, too much CPU but not enough memory, or the reverse.

Goal: right-size the resource mix rather than simply adding or removing capacity.

Decommission

Recommended when a resource appears unused or no longer required.

Indicators include systems that have been powered off for an extended period.

Goal: eliminate unnecessary resources to reduce cost, risk, and operational overhead.

Each resource in Insights is assigned a category, shown in the Category column.

Reduce costs

The resource has more capacity than it needs, or appears to be unused.

A Downsize or Decommission is recommended, and the estimated savings are shown in the Annual cost and Delta columns.

Reduce risks

The resource needs more capacity than it currently has.

An Upsize is recommended. This may increase cost but improves performance and stability.

Optimal

Usage is within the configured thresholds and no change is needed.

A resource can also be shown as Optimal when a change would only make a small difference in cost.

Constrained

Usage is outside the configured thresholds, but Insights cannot recommend a different size that would safely address it.

A common example is a VM with low CPU usage most of the time but short spikes in activity. Reducing its vCPUs would leave it unable to handle those spikes, so the current size is kept.

Constrained does not necessarily mean the resource is short on capacity. No cost estimate is shown because no change is recommended.

Excluded

The resource was not evaluated.

This can happen when there is not yet enough usage data, the VM was only recently powered off, or the VM has a role that is not suited to resizing, such as a Kubernetes node, a system VM, or a desktop operating system.

For data center VMs, Insights bases the recommended memory size on Memory used. This is the same metric shown in the Memory used column of the Entities view and Excel export. For VMware VMs, it's the guest's memory needed as reported by VMware Tools, or memory consumed if VMware Tools isn't running.

The recommended size is calculated as follows:

The minimum recommended size is 512 MiB.

Memory is only recalculated when a VM falls outside its target range for CPU or memory. A VM with low CPU usage can therefore also receive a new memory size, even if its memory usage is within range.

Example

A VM has 4096 MiB of memory configured, and its highest Memory used over the last 14 days was 3359 MiB:

3359 MiB ÷ 0.96 = 3499 MiB, rounded up to the next 512 MiB = 3584 MiB

Estimating from the Excel export

You can estimate the recommendation from the Entities Excel export using:

Memory max % × Memory (MiB) ÷ 0.96, rounded up as described above

Memory max % covers only the export's selected time range, while Insights looks at up to the last 14 days, so the two may not match exactly. If the IT Environment owner has changed the thresholds, use their upper memory threshold instead of 0.96.

Insights recommendations are driven by thresholds that define what is considered overutilized, underutilized, or unused. These thresholds are configurable. Only the user who created the IT Environment can adjust these thresholds. They are managed from the settings cog within the IT Environment. Workspace users cannot modify thresholds. They can view Insights and the resulting recommendations, but any changes to threshold values must be made by the IT Environment owner.