Skip to content

title: CloudSlash - Cost Optimization Reference description: You have 847 waste items costing $14,200/month. Focus on the top 20: they account for 80% of it. Here is exactly how CloudSlash gets you there. The engine di...

Cost Optimization

You have 847 waste items costing $14,200/month. Focus on the top 20: they account for 80% of it. Here is exactly how CloudSlash gets you there.

Prerequisites

  • CloudSlash installed (cs doctor passes)
  • AWS credentials configured (aws sts get-caller-identity succeeds)
  • AWS plugin installed: cs plugin install aws

Idle Resources

Target First

cs scan --region us-east-1

The engine discovers all resources in your account, builds the Universal Graph, runs heuristics, and outputs a table of findings:

┌─────────────────────────────────────────────────────────────────────────────┐
│ VERDICT │ RESOURCE                      │ COST/MO │ REASON                │
├─────────────────────────────────────────────────────────────────────────────┤
│ DELETE  │ i-0abc123 (m5.xlarge)         │ $140.16 │ idle_instance (45d)   │
│ DELETE  │ vol-0def456 (100GB gp2)       │ $10.00  │ unattached_volume     │
│ FLAG    │ lambda-stale-processor        │ $0.00   │ stale_lambda (90d)    │
│ FLAG    │ nat-0ghi789                   │ $32.40  │ hollow_nat_gateway    │
│ RISK    │ i-0jkl012 (c5.2xlarge)       │ $245.28 │ spot_termination_risk │
└─────────────────────────────────────────────────────────────────────────────┘

Waste: 4 resources │ Projected savings: $182.56/mo

Web Dashboard

Explore in the

Start the daemon to get the full visual experience:

cs daemon --region us-east-1

Open http://localhost:8080 and navigate to the Insights tab. Here you see:

  • Cost breakdown by provider, service, and region
  • Top waste items ranked by monthly cost
  • Heuristic confidence scores for each finding

Tune Detection Sensitivity

Open Settings (gear icon) → Heuristic Sliders. Adjust thresholds for your environment:

  • Idle instance threshold: Days of low CPU before flagging (default: 14 days)
  • Volume detach threshold: Days unattached before flagging (default: 14 days)
  • Lambda staleness: Days since last invocation (default: 90 days)

Changes apply immediately to the next scan cycle.

Remediations

Review and Approve

Navigate to the s.a.u Pipeline tab. Resources flow through the Kanban stages:

  1. Flagged: Heuristics identified waste
  2. Analyzed: Oracle verified blast radius
  3. s.a.u Gate: Reversibility check passed
  4. Pending: Awaiting your approval

Click a resource to see its Terraform plan diff and blast radius. When you are confident, click Approve or drag it to the next stage.

Dry Run First

For your first remediations, use the Web Dashboard to review each resource individually. The blast radius view shows all dependent resources that could be affected.

Track Savings

Over Time

cs trend              # Last 30 days
cs trend --days=90    # Quarterly view

The trend chart shows waste count and projected savings over time. Use it to demonstrate ROI to stakeholders.

Generate Executive Reports

cs report --format=html --output=cost-optimization-report.html

The report includes: executive summary, provider breakdown, top 10 waste items, policy compliance status, and cost trend charts. Share it with your finance team.

What's Next