Scanner Settings
Configure how aggressively StratoLens recommends downsizing and how long a new VM must run before its first recommendation appears. StratoLens is read-only and never modifies your VMs. Open Settings > Scanner from the left sidebar and expand VM Sizing Recommendations.
Save permission required
Saving requires the "Configure Scanner Settings" permission. Without it, the section is read-only and Save Changes stays disabled. Read access to the section itself is still available so you can see the current configuration.
Sizing Aggressiveness
Aggressiveness controls the target maximum utilization the engine sizes against. Higher aggressiveness means tighter sizing with less buffer for workload spikes, yielding larger savings.
- Conservative
- 70% target utilization. Largest headroom for workload spikes, smallest savings.
- Balanced
- 80% target utilization. Default. Moderate headroom and good savings.
- Aggressive
- 90% target utilization. Tightest sizing, maximum savings, least buffer.
Defaults
- Aggressiveness
- Balanced (80%)
Start with Balanced
Balanced is the recommended starting point. Move to Aggressive only when you have buffer in your workload-spike tolerance — for example, in non-production environments or for workloads that scale horizontally and can absorb a tighter ceiling.
No per-VM override
Aggressiveness applies to every recommendation in your environment. To leave a specific VM alone, hide its recommendation from the Recommendations page rather than changing aggressiveness.
Changing aggressiveness regenerates every recommendation
On the next scan, every VM's recommendation is recomputed against the new utilization ceiling. Existing slot type badges and savings figures may all change.
Minimum Data Window
The minimum number of days of performance data a VM must accumulate before its first recommendation is created. Once a recommendation exists, it continues to be updated regardless of this threshold. Valid range is 0 to 365 days. Set to 0 to include every VM.
Defaults
- Minimum Data Window
- 7 days
Short windows lower confidence
Recommendations from very short data windows have lower confidence scores because data sufficiency is one of the inputs to confidence. Use them as directional signals, not commitments, until more history has accumulated.
Troubleshooting
I changed aggressiveness but my recommendations look the same
What to check
Aggressiveness is read at scan time. Wait for the next scheduled scan or trigger a manual scan. Recommendations are regenerated during the scan, not on save.
Save Changes is greyed out
What to check
Either there are no pending changes to save, or your role doesn't include the "Configure Scanner Settings" permission. Hover the button for the tooltip explanation.
I want recommendations for a VM that's only been running for 3 days
What to check
Lower the Minimum Data Window to 3 (or 0 to include every VM), save, then wait for the next scan. Note that recommendations from very short windows have lower confidence — they're still useful for directional sizing, but treat them as preliminary.