Skip to main content

cloud-cost-models

AWS/GCP/Azure cost models, sizing guides, Reserved Instance/Savings Plan strategies, and FinOps framework guide. Use this skill for 'cloud costs', 'cost optimization', 'FinOps', 'instance sizing', 'Savings Plan', 'reserved instances', 'Spot instances', 'cost estimation', and other infrastructure cost-related decisions. Enhances the cost analysis capabilities of cost-optimizer. Note: actual resource provisioning and billing configuration are outside the scope of this skill.

Source facts

Repository
revfactory/harness-100
Last source activity
March 22, 2026 at 22:41
Detected SKILL.md language
English
Stars
1,283
Forks
444

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

Showing SKILL.md

SKILL.md
Source instructions · Read-only preview
name
cloud-cost-models
description
AWS/GCP/Azure cost models, sizing guides, Reserved Instance/Savings Plan strategies, and FinOps framework guide. Use this skill for 'cloud costs', 'cost optimization', 'FinOps', 'instance sizing', 'Savings Plan', 'reserved instances', 'Spot instances', 'cost estimation', and other infrastructure cost-related decisions. Enhances the cost analysis capabilities of cost-optimizer. Note: actual resource provisioning and billing configuration are outside the scope of this skill.
# Cloud Cost Models — Cloud Cost Model and FinOps Guide A practical guide for accurately estimating and optimizing cloud infrastructure costs. ## AWS Core Service Cost Models ### EC2 Purchase Option Comparison (ap-northeast-2 basis) | Option | Discount | Commitment | Flexibility | Suitable Workload | |--------|---------|-----------|------------|-------------------| | On-Demand | 0% | None | Maximum | Variable workloads, testing | | Savings Plan (Compute) | ~30% | 1 year | Instance changeable | Stable baseline load | | Savings Plan (EC2) | ~40% | 1 year | Within family | Predictable workloads | | Reserved Instance | ~40% | 1yr/3yr | Limited | 24/7 servers | | Spot Instance | ~70% | None | Interruptible | Batch, CI/CD, ML | ### Sizing Decision Table ``` CPU utilization < 20% sustained -> Downsizing recommended CPU utilization > 80% sustained -> Upsize or scale out Memory utilization < 30% -> Memory-optimized instance unnecessary Workload type -> Instance family: +-- General purpose: t3, m6i (CPU/memory balanced) +-- Compute intensive: c6i (API servers, computation) +-- Memory intensive: r6i (cache, in-memory DB) +-- Storage intensive: i3, d3 (databases) +-- GPU: p4d, g5 (ML training/inference) ``` ### RDS Cost Components ``` Monthly cost = Instance cost + Storage + I/O + Backup + Transfer Example: db.r6g.xlarge, Multi-AZ, 500GB gp3 +-- Instance: $0.54/h x 730h x 2(Multi-AZ) = $788.40 +-- Storage: 500GB x $0.138/GB = $69.00 +-- Backup (excess): 100GB x $0.095/GB = $9.50 +-- Monthly total: ~$866.90 ``` ## Per-environment Cost Optimization Strategies | Environment | Strategy | Expected Savings | |------------|----------|-----------------| | **dev** | Business hours only (12h/day, 5 days/week) = 36% uptime | ~60% savings | | **staging** | Same as dev + smaller instances | ~70% savings | | **prod** | Savings Plan + Right-sizing + Auto-scaling | ~40% savings | ### Dev Environment Scheduling ```hcl # Lambda for auto start/stop of dev environment resource "aws_cloudwatch_event_rule" "start_dev" { schedule_expression = "cron(0 0 ? * MON-FRI *)" # Weekdays 09:00 KST } resource "aws_cloudwatch_event_rule" "stop_dev" { schedule_expression = "cron(0 12 ? * MON-FRI *)" # Weekdays 21:00 KST } ``` ## Cost Estimation Template ```markdown # Monthly Cost Estimate ## Architecture Summary | Component | Specs | Quantity | ## Per-environment Costs | Service | Dev | Staging | Prod | Total | |---------|-----|---------|------|-------| | EC2/ECS | | | | | | RDS | | | | | | ElastiCache | | | | | | ALB | | | | | | S3 | | | | | | CloudWatch | | | | | | Data Transfer | | | | | | **Subtotal** | | | | | ## Optimization Applied | Optimization | Savings | |-------------|---------| | Savings Plan (1 year) | | | Dev env scheduling | | | Right-sizing | | | **Total savings** | | ## Final Cost - Before optimization: $N/month - After optimization: $N/month - Savings rate: N% ``` ## FinOps Maturity Model | Phase | Activities | Tools | |-------|-----------|-------| | **Inform** | Cost visibility, tagging, allocation | AWS Cost Explorer, tag policies | | **Optimize** | Right-sizing, reservations, spot | Compute Optimizer, Trusted Advisor | | **Operate** | Budget alerts, automated actions, governance | AWS Budgets, Lambda auto-cleanup |
View on GitHub