Waste Findings Categories and Priorities
Who this is for
Anyone who wants to reduce cloud spend by identifying and eliminating wasted resources.
What you will complete
Understand each category of waste CloudAIPilot detects, how findings are prioritized, and how to act on them.
Before you begin
- FinOps billing data must be connected.
- Navigate to FinOps → Waste card.
Waste categories
CloudAIPilot detects seven categories of cloud waste:
1. Right-size (highest savings potential)
What it is: A server that is significantly over-provisioned — using only 10–20% of its CPU and memory consistently, while being billed for a much larger instance.
Typical saving: 40–70% of that server's cost by moving to a smaller instance type.
How to act: Review the recommendation details (see KB-08-09). Verify the low utilization is not due to burst traffic patterns. If confirmed, resize the server to the recommended instance type.
2. Idle server
What it is: A server that has been powered on but shows near-zero activity (CPU, network, requests) for 7+ days. Likely a forgotten development or testing server.
Typical saving: 100% of that server's cost by stopping or terminating it.
How to act: Confirm no active workloads depend on it. Check with the team who last used it. Stop or archive if confirmed idle.
3. Orphan volume
What it is: A cloud storage volume (EBS volume, GCP persistent disk, Azure managed disk) that is no longer attached to any server. The volume is being billed even though no server uses it.
Typical saving: Full cost of the volume (typically $0.10/GB/month, adds up for large volumes).
How to act: Verify no server needs the volume. Take a snapshot for safety, then delete the volume.
4. Orphan snapshot
What it is: An old backup snapshot that is no longer needed — for example, snapshots from servers that have been deleted, or snapshots older than your retention policy.
Typical saving: Storage cost of old snapshots (can be significant for large disks over time).
How to act: Review the snapshot age and source. Delete snapshots that exceed your retention window.
5. Unattached IP
What it is: A static/elastic IP address reserved in your cloud account that is not attached to any running server. Most cloud providers charge for unattached reserved IPs.
Typical saving: $3–18/month per unattached IP (varies by provider and region).
How to act: Release the IP in your cloud provider console, or attach it to a server that needs a static IP.
6. Expensive zone
What it is: A server or resource running in a cloud region that is significantly more expensive than equivalent regions. Some regions cost 20–40% more for the same instance type.
Typical saving: 15–30% by migrating to a cheaper equivalent region (if latency requirements allow).
How to act: Verify latency requirements before acting. If a cheaper region is acceptable, plan a migration.
7. Commitment opportunity
What it is: Workloads running on on-demand pricing that would be cheaper with Reserved Instances (AWS), Committed Use Discounts (GCP), or Reserved VM Instances (Azure).
Typical saving: 30–60% of compute costs for stable, predictable workloads.
How to act: Review the workload's utilization history. If consistently above 70% utilization for 6+ months, evaluate commitment purchases directly in your cloud provider console.
Priority levels
Each waste finding is assigned a priority:
| Priority | Criteria |
|---|---|
| High | Estimated monthly saving > $50, or resource is clearly unused |
| Medium | Estimated monthly saving $10–$50, or resource is likely unused |
| Low | Estimated monthly saving < $10, or usage pattern is ambiguous |
Focus on High priority findings first — they deliver the most value for the least effort.