v0.1.0 to v0.1.1
25 added, 81 removed. Audit A to A.
---
name: azure-waf-cost-optimization-review
description: "Review Azure workload cost posture against the Well-Architected Framework Cost Optimization pillar: cost modeling, rightsizing, reservations, hybrid benefit, storage lifecycle, and idle resource elimination."
allowed-tools: Read Grep Glob
metadata:
- author: "github: Raishin"
- version: "0.1.0"
- updated: "2026-05-09"
+ author: github: Raishin
+ version: 0.1.1
+ updated: "2026-06-04"
category: finops
---
- # Azure WAF Cost Optimization Review
-
- The Azure Well-Architected Framework Cost Optimization pillar focuses on designing workloads that maximize the value delivered while minimizing unnecessary Azure spending.
-
- ## 5 Cost Optimization Design Principles
-
- 1. **Develop a cost model** — model costs before deployment; estimate spend based on requirements; track cost KPIs
- 2. **Design with a cost-efficiency mindset** — favor managed services, consumption-based pricing, and right-sized resources
- 3. **Design for usage optimization** — enable auto-scaling; schedule workloads; use spot/low-priority VMs for fault-tolerant jobs
- 4. **Design for rate optimization** — commit to consistent usage via Azure Reservations and Savings Plans; leverage hybrid benefits
- 5. **Monitor and optimize over time** — track spend continuously; act on cost recommendations; retire underutilized resources
-
- ## Azure Cost Tools
-
- - **Microsoft Cost Management + Billing** — cost analysis, budgets, cost alerts, export to storage for Parquet/Athena analysis
- - **Azure Advisor** — cost recommendations: resize/shutdown VMs, buy reservations, delete unused disks
- - **Azure Monitor + cost alerts** — alert on budget thresholds
- - **Azure Pricing Calculator** — pre-deployment cost estimation
- - **Azure Cost Management Exports** — raw billing data to Azure Storage for Power BI / Excel analysis
- - **Azure Carbon Optimization** (preview) — sustainability tracking for carbon footprint reporting
-
- ## Cost Optimization Areas
-
- ### Compute
- - Right-size VMs using Azure Advisor; use Burstable VMs (B-series) for dev/test
- - Spot VMs — up to 90% off list price for interruptible workloads (CI/CD, ML training, batch)
- - Azure Batch Spot nodes — bursty HPC at spot pricing
- - Auto-shutdown dev VMs outside business hours via Azure DevTest Labs or scheduled stop
-
- ### Reservations and Savings Plans
- - **1yr/3yr Azure Reservations** — VMs, SQL, Cosmos DB, App Service, AKS nodes; committed capacity discount
- - **Azure Savings Plans for Compute** — covers VMs, VMSS, AKS, App Service, Functions across regions, SKU families, and sizes; no instance flexibility restrictions
-
- ### Azure Hybrid Benefit
- - **Windows Server** — up to 40% off VM pricing using existing Software Assurance licenses
- - **SQL Server** — up to 55% off Azure SQL Database and SQL Managed Instance
- - **Dev/Test pricing** — reduced rates for Visual Studio subscribers on eligible services
-
- ### Storage
- - **Azure Blob lifecycle management** — tiered transition: Hot → Cool → Cold → Archive based on last access time
- - **Storage Reserved Capacity** — commit to blob storage capacity for up to 3 years
- - LRS vs GRS vs GZRS selection — match redundancy to actual DR requirements; avoid over-replication
-
- ### Networking
- - Azure Front Door vs Application Gateway vs Load Balancer cost tradeoffs
- - ExpressRoute vs VPN Gateway pricing model differences
- - Egress data transfer costs — VNet peering is free; egress to internet is billed per GB
-
- ### Governance
- - Management Group + Subscription tagging policy — enforce required tags via Azure Policy deny
- - Cost allocation tags — chargeback/showback reporting by team, application, environment, cost-center
- - Azure Policy to enforce allowed VM SKUs — prevent accidental high-cost SKU deployment
+ # Azure WAF Cost Optimization
- ## Assessment Questions
+ ## Purpose
- - How do you model and predict cloud costs for your workload?
- - How do you ensure resources are correctly sized for the workload?
- - How do you use Azure pricing models (reservations, savings plans, spot) to reduce costs?
- - How do you monitor and act on cost anomalies?
- - How do you retire unused or underutilized resources?
- - How do you allocate costs to teams or business units?
- - How do you evaluate the cost impact of architectural decisions?
+ Act as a ruthless Azure reviewer for azure waf cost optimization work. Stop broad, vague, or unverified recommendations before they become production risk.
- ## Validation Checklist
+ ## Lean operating rules
- - [ ] Azure Cost Management budgets configured per subscription with email alerts at 80% and 100%
- - [ ] 100% of resources tagged with required tags (env, team, app, cost-center) via Azure Policy
- - [ ] Azure Advisor cost recommendations reviewed and actioned monthly
- - [ ] Azure Reservations or Savings Plans covering ≥70% of steady-state compute (VMs, AKS nodes, SQL)
- - [ ] Azure Hybrid Benefit enabled for eligible Windows Server and SQL Server workloads
- - [ ] Spot VMs or Azure Batch Spot nodes used for CI/CD, ML training, and dev/test workloads
- - [ ] Blob Storage lifecycle policies active for all storage accounts with data older than 30 days
- - [ ] Unused resources identified and decommissioned monthly: unattached disks, unused public IPs, empty resource groups, stopped VMs >7 days
- - [ ] Dev/test VMs auto-shutdown scheduled outside business hours
- - [ ] Cost allocation report generated monthly for chargeback/showback
+ - Prefer Microsoft Learn documentation through the user's configured documentation MCP, then sampled read-only Azure evidence when available, then sanitized user evidence.
+ - Separate confirmed facts from inference. If state was not queried or shown, say so.
+ - Challenge broad access, broad scope, destructive changes, billing-impacting actions, and hand-wavy production claims.
+ - Keep the answer scoped, reversible where possible, least-privilege, and explicit about blockers or unknowns.
+ - Never ask the user to paste credentials, tokens, secrets, tenant IDs, subscription IDs, resource IDs, customer data, private keys, or raw incident payloads.
- ## Response Shape
+ ## References
- Cost visibility and tooling assessment → tagging compliance → reservation/savings plan coverage → rightsizing opportunities → hybrid benefit and spot adoption → storage lifecycle → idle resource inventory → prioritized savings actions
+ Load these only when needed:
- ## Official Documentation
+ - [Azure WAF Cost Optimization Operations](references/waf-cost-optimization-operations.md) — use for current service behavior, common failure modes, hard design rules, verification targets, and push-back conditions.
+ - [Safety checklist](references/safety-checklist.md) — use for evidence labels, risk gates, mutation boundaries, approval rules, credential boundaries, and current-state caveats.
+ - [MCP and evidence path](references/mcp-and-evidence.md) — use when choosing documentation-based evidence, sampled read-only evidence, or sanitized user evidence.
+ - [Workflow and output contract](references/workflow-and-output.md) — use when executing the full review, applying stress checks, or formatting the final answer.
+ - [Official sources](references/official-sources.md) — use when you need the detailed Microsoft documentation list or source notes.
- - https://learn.microsoft.com/azure/well-architected/cost-optimization/
- - https://learn.microsoft.com/azure/cost-management-billing/
+ ## Response minimum
- ## Security Notes
+ Return, at minimum:
- Read-only advisory. Do not cancel Reservations, delete resources, or modify billing configurations without explicit approval and resource inventory confirmation.
+ - the scoped target and evidence level,
+ - the main risks or control gaps,
+ - the safest next actions,
+ - the assumptions or blockers that prevent stronger conclusions.