Identifying and Reviewing Idle Virtual Machines in Aria Operations

Products

VCF Operations VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

You are seeking guidance on how VMware Aria Operations defines an idle virtual machine (VM) and the recommended frequency for reviewing these resources to optimize your environment.

Environment

VMware Aria Operations

Resolution

This guidance is provided to help you understand the technical criteria for idleness and establish an operational strategy for resource reclamation.

How Aria Operations Defines an Idle VM

By default, you can identify a virtual machine as idle if its resource consumption stays below a specific threshold for an entire day. A VM is considered idle if its CPU consumption is less than 100 MHz for 100% of the time during each 24-hour period.

You can modify these thresholds by navigating to Administration > Global Settings > Resource Reclamation. Here, you can adjust the MHz threshold and the percentage of time the VM must stay below that threshold to be flagged.

Under Idle VM Management, you can configure two additional parameters:

Idle Period: The number of consecutive days a VM must meet the idleness criteria before the system classifies it as "reclaimable."
Exclusion Period: The number of days since a VM was provisioned before the system considers it for idleness. This prevents you from flagging new VMs while they are still being configured.

Recommendation for Review Frequency

While there is no single mandatory frequency, Broadcom suggests the following best practices:

System Update Frequency (Daily): Aria Operations runs its orphaned disk reclamation sync cycle daily (defaulting to 8:00 PM local time) and performs cost calculations shortly after (9:00 PM). The recommendations you see in the UI refresh daily based on this data.
Managerial/Operational Review (Weekly): For most IT managers, a weekly review of the Capacity > Reclaim dashboard is recommended. This aligns with standard change management windows and allows for a "cooling off" period to verify if a VM is truly abandoned or just temporarily quiet.
Automated Reclamation: If you use Automation Central, you can schedule reclamation jobs to run automatically (e.g., monthly) for low-risk items like snapshots or orphaned disks. However, manual review is still recommended for powered-on "Idle" VMs.

Important Notes

vCLS VMs: vSphere Cluster Services (vCLS) VMs may be detected as idle by the capacity optimization workflow. Do not reclaim or delete these VMs, as they are critical for cluster functions like DRS and HA.
Data Integrity: Ensure that you schedule the Orphaned Disks Collection time before the Cost Calculation time in Global Settings so that potential savings are calculated based on the most recent list of reclaimable disks.
Reclamation Order: It is often safer for you to prioritize reclaiming Powered Off VMs and Snapshots before addressing Idle VMs, as the latter are still running and may belong to seasonal or intermittent workloads.

Additional Information

Reclamation Settings