🔍⌘K

Start typing to search docs.

Cost Management Runbook

1.0.0

Monitor Supabase costs and optimize shared infra.

On this page
  1. Cost Management Runbook
  2. 🎯 Overview
  3. 💰 Cost Monitoring
  4. AWS Cost Monitoring
  5. Daily Cost Check
  6. Weekly Cost Analysis
  7. GCP Cost Monitoring
  8. Daily Cost Check
  9. Weekly Cost Analysis
  10. 📊 Budget Management
  11. AWS Budget Setup
  12. GCP Budget Setup
  13. 🔧 Cost Optimization
  14. AWS Cost Optimization
  15. RDS Optimization
  16. S3 Optimization
  17. GCP Cost Optimization
  18. Cloud SQL Optimization
  19. Cloud Storage Optimization
  20. 🚨 Emergency Cost Controls
  21. AWS Emergency Shutdown
  22. GCP Emergency Shutdown
  23. 📈 Cost Reporting
  24. Monthly Cost Report
  25. Cost Optimization Recommendations

Cost Management Runbook

This runbook provides comprehensive procedures for monitoring, optimizing, and controlling infrastructure costs across AWS and GCP deployments.

🎯 Overview

This runbook covers:

  • Cost monitoring and alerting
  • Budget management
  • Cost optimization strategies
  • Regular cost reviews
  • Emergency cost controls

💰 Cost Monitoring

AWS Cost Monitoring

Daily Cost Check

#!/bin/bash
# aws-daily-cost-check.sh

echo "💰 AWS Daily Cost Check"
echo "======================"

# Get yesterday's costs
YESTERDAY=$(date -d "yesterday" +%Y-%m-%d)
TODAY=$(date +%Y-%m-%d)

echo "📊 Costs for $YESTERDAY:"

aws ce get-cost-and-usage \
  --time-period Start=$YESTERDAY,End=$TODAY \
  --granularity DAILY \
  --metrics BlendedCost \
  --group-by Type=DIMENSION,Key=SERVICE \
  --query 'ResultsByTime[0].Groups[?Metrics.BlendedCost.Amount>`0.01`].[Keys[0],Metrics.BlendedCost.Amount]' \
  --output table

# Get month-to-date costs
MONTH_START=$(date +%Y-%m-01)

echo ""
echo "📈 Month-to-date costs:"

aws ce get-cost-and-usage \
  --time-period Start=$MONTH_START,End=$TODAY \
  --granularity MONTHLY \
  --metrics BlendedCost \
  --query 'ResultsByTime[0].Total.BlendedCost.Amount' \
  --output text

Weekly Cost Analysis

#!/bin/bash
# aws-weekly-cost-analysis.sh

echo "📊 AWS Weekly Cost Analysis"
echo "=========================="

# Get last 7 days of costs
WEEK_AGO=$(date -d "7 days ago" +%Y-%m-%d)
TODAY=$(date +%Y-%m-%d)

# Cost by service
echo "💸 Top services by cost (last 7 days):"
aws ce get-cost-and-usage \
  --time-period Start=$WEEK_AGO,End=$TODAY \
  --granularity DAILY \
  --metrics BlendedCost \
  --group-by Type=DIMENSION,Key=SERVICE \
  --query 'ResultsByTime[].Groups[?Metrics.BlendedCost.Amount>`1.00`].[Keys[0],Metrics.BlendedCost.Amount]' \
  --output table | head -20

# Cost by resource
echo ""
echo "🏷️ Top resources by cost (last 7 days):"
aws ce get-cost-and-usage \
  --time-period Start=$WEEK_AGO,End=$TODAY \
  --granularity DAILY \
  --metrics BlendedCost \
  --group-by Type=DIMENSION,Key=RESOURCE_ID \
  --query 'ResultsByTime[].Groups[?Metrics.BlendedCost.Amount>`5.00`].[Keys[0],Metrics.BlendedCost.Amount]' \
  --output table | head -10

GCP Cost Monitoring

Daily Cost Check

#!/bin/bash
# gcp-daily-cost-check.sh

echo "💰 GCP Daily Cost Check"
echo "======================"

PROJECT_ID=$(gcloud config get-value project)
YESTERDAY=$(date -d "yesterday" +%Y-%m-%d)

echo "📊 Costs for $YESTERDAY (Project: $PROJECT_ID):"

# Get billing data (requires billing export to BigQuery)
bq query --use_legacy_sql=false \
"SELECT
  service.description as service,
  ROUND(SUM(cost), 2) as cost_usd
FROM \`$PROJECT_ID.billing.gcp_billing_export_v1_*\`
WHERE DATE(usage_start_time) = '$YESTERDAY'
  AND cost > 0.01
GROUP BY service
ORDER BY cost_usd DESC
LIMIT 10"

Weekly Cost Analysis

#!/bin/bash
# gcp-weekly-cost-analysis.sh

echo "📊 GCP Weekly Cost Analysis"
echo "=========================="

PROJECT_ID=$(gcloud config get-value project)
WEEK_AGO=$(date -d "7 days ago" +%Y-%m-%d)
TODAY=$(date +%Y-%m-%d)

echo "💸 Top services by cost (last 7 days):"

bq query --use_legacy_sql=false \
"SELECT
  service.description as service,
  ROUND(SUM(cost), 2) as cost_usd,
  COUNT(*) as usage_records
FROM \`$PROJECT_ID.billing.gcp_billing_export_v1_*\`
WHERE DATE(usage_start_time) BETWEEN '$WEEK_AGO' AND '$TODAY'
  AND cost > 0
GROUP BY service
ORDER BY cost_usd DESC
LIMIT 15"

📊 Budget Management

AWS Budget Setup

#!/bin/bash
# setup-aws-budgets.sh

echo "💰 Setting up AWS Budgets"
echo "========================"

# Monthly budget for development environment
cat > dev-budget.json << EOF
{
  "BudgetName": "app-factory-development-monthly",
  "BudgetLimit": {
    "Amount": "200",
    "Unit": "USD"
  },
  "TimeUnit": "MONTHLY",
  "TimePeriod": {
    "Start": "$(date +%Y-%m-01)",
    "End": "2030-12-31"
  },
  "CostFilters": {
    "TagKey": ["Environment"],
    "TagValue": ["development"]
  },
  "BudgetType": "COST"
}
EOF

# Create budget
aws budgets create-budget \
  --account-id $(aws sts get-caller-identity --query Account --output text) \
  --budget file://dev-budget.json

# Production budget
cat > prod-budget.json << EOF
{
  "BudgetName": "app-factory-production-monthly",
  "BudgetLimit": {
    "Amount": "1000",
    "Unit": "USD"
  },
  "TimeUnit": "MONTHLY",
  "TimePeriod": {
    "Start": "$(date +%Y-%m-01)",
    "End": "2030-12-31"
  },
  "CostFilters": {
    "TagKey": ["Environment"],
    "TagValue": ["production"]
  },
  "BudgetType": "COST"
}
EOF

aws budgets create-budget \
  --account-id $(aws sts get-caller-identity --query Account --output text) \
  --budget file://prod-budget.json

echo "✅ AWS budgets created successfully"

GCP Budget Setup

#!/bin/bash
# setup-gcp-budgets.sh

echo "💰 Setting up GCP Budgets"
echo "========================"

PROJECT_ID=$(gcloud config get-value project)
BILLING_ACCOUNT=$(gcloud billing projects describe $PROJECT_ID --format="value(billingAccountName)" | cut -d'/' -f2)

# Development environment budget
gcloud billing budgets create \
  --billing-account=$BILLING_ACCOUNT \
  --display-name="App Factory Development Monthly" \
  --budget-amount=200USD \
  --threshold-rule=percent=50,basis=current-spend \
  --threshold-rule=percent=90,basis=current-spend \
  --threshold-rule=percent=100,basis=current-spend \
  --filter-projects=$PROJECT_ID \
  --filter-labels=environment=development

# Production environment budget
gcloud billing budgets create \
  --billing-account=$BILLING_ACCOUNT \
  --display-name="App Factory Production Monthly" \
  --budget-amount=1000USD \
  --threshold-rule=percent=50,basis=current-spend \
  --threshold-rule=percent=90,basis=current-spend \
  --threshold-rule=percent=100,basis=current-spend \
  --filter-projects=$PROJECT_ID \
  --filter-labels=environment=production

echo "✅ GCP budgets created successfully"

🔧 Cost Optimization

AWS Cost Optimization

RDS Optimization

#!/bin/bash
# optimize-aws-rds.sh

echo "🗄️ AWS RDS Cost Optimization"
echo "============================"

# Find underutilized RDS instances
echo "📊 Checking RDS utilization..."

aws rds describe-db-instances \
  --query 'DBInstances[?DBInstanceStatus==`available`].[DBInstanceIdentifier,DBInstanceClass,Engine,AllocatedStorage]' \
  --output table

# Get CloudWatch metrics for CPU utilization
for instance in $(aws rds describe-db-instances --query 'DBInstances[?DBInstanceStatus==`available`].DBInstanceIdentifier' --output text); do
    echo "📈 CPU utilization for $instance (last 7 days):"
    
    aws cloudwatch get-metric-statistics \
      --namespace AWS/RDS \
      --metric-name CPUUtilization \
      --dimensions Name=DBInstanceIdentifier,Value=$instance \
      --start-time $(date -d "7 days ago" --iso-8601) \
      --end-time $(date --iso-8601) \
      --period 86400 \
      --statistics Average \
      --query 'Datapoints[].Average' \
      --output text | awk '{sum+=$1; count++} END {if(count>0) printf "Average: %.2f%%\n", sum/count}'
done

S3 Optimization

#!/bin/bash
# optimize-aws-s3.sh

echo "🪣 AWS S3 Cost Optimization"
echo "=========================="

# Find large S3 buckets
echo "📊 S3 bucket sizes:"
aws s3 ls | while read -r line; do
    bucket=$(echo $line | awk '{print $3}')
    if [ ! -z "$bucket" ]; then
        size=$(aws s3 ls s3://$bucket --recursive --summarize | grep "Total Size" | awk '{print $3}')
        if [ ! -z "$size" ] && [ "$size" -gt 1000000000 ]; then  # > 1GB
            echo "$bucket: $(echo $size | awk '{printf "%.2f GB", $1/1024/1024/1024}')"
        fi
    fi
done

# Check lifecycle policies
echo ""
echo "🔄 Checking lifecycle policies:"
aws s3api list-buckets --query 'Buckets[].Name' --output text | while read bucket; do
    lifecycle=$(aws s3api get-bucket-lifecycle-configuration --bucket $bucket 2>/dev/null)
    if [ $? -eq 0 ]; then
        echo "✅ $bucket has lifecycle policy"
    else
        echo "❌ $bucket missing lifecycle policy"
    fi
done

GCP Cost Optimization

Cloud SQL Optimization

#!/bin/bash
# optimize-gcp-cloudsql.sh

echo "🗄️ GCP Cloud SQL Cost Optimization"
echo "=================================="

# List Cloud SQL instances
echo "📊 Cloud SQL instances:"
gcloud sql instances list --format="table(name,tier,region,status)"

# Check CPU utilization
for instance in $(gcloud sql instances list --format="value(name)"); do
    echo ""
    echo "📈 Checking utilization for $instance..."
    
    # Get recent CPU metrics (requires monitoring API)
    gcloud logging read "resource.type=cloud_sql_database AND resource.labels.database_id=$instance" \
      --limit=10 \
      --format="value(timestamp,jsonPayload.message)" 2>/dev/null || echo "No recent logs found"
done

Cloud Storage Optimization

#!/bin/bash
# optimize-gcp-storage.sh

echo "🪣 GCP Cloud Storage Cost Optimization"
echo "====================================="

# List storage buckets with sizes
echo "📊 Storage bucket analysis:"
gsutil ls -L -b gs://* | grep -E "(gs://|Storage class|Total size)" | \
  awk '/gs:\/\// {bucket=$0} /Storage class/ {class=$3} /Total size/ {size=$3; print bucket, class, size}'

# Check lifecycle policies
echo ""
echo "🔄 Checking lifecycle policies:"
gsutil ls | while read bucket; do
    lifecycle=$(gsutil lifecycle get $bucket 2>/dev/null)
    if [ $? -eq 0 ]; then
        echo "✅ $bucket has lifecycle policy"
    else
        echo "❌ $bucket missing lifecycle policy"
    fi
done

🚨 Emergency Cost Controls

AWS Emergency Shutdown

#!/bin/bash
# aws-emergency-shutdown.sh

echo "🚨 AWS Emergency Cost Control"
echo "============================"

read -p "⚠️ This will shut down non-essential resources. Continue? (y/N) " -n 1 -r
echo
if [[ ! $REPLY =~ ^[Yy]$ ]]; then
    echo "❌ Emergency shutdown cancelled"
    exit 1
fi

# Stop non-production RDS instances
echo "🛑 Stopping development RDS instances..."
aws rds describe-db-instances \
  --query 'DBInstances[?contains(Tags[?Key==`Environment`].Value, `development`) && DBInstanceStatus==`available`].DBInstanceIdentifier' \
  --output text | while read instance; do
    if [ ! -z "$instance" ]; then
        echo "Stopping $instance..."
        aws rds stop-db-instance --db-instance-identifier $instance
    fi
done

# Stop non-production EC2 instances
echo "🛑 Stopping development EC2 instances..."
aws ec2 describe-instances \
  --filters "Name=tag:Environment,Values=development" "Name=instance-state-name,Values=running" \
  --query 'Reservations[].Instances[].InstanceId' \
  --output text | while read instance; do
    if [ ! -z "$instance" ]; then
        echo "Stopping $instance..."
        aws ec2 stop-instances --instance-ids $instance
    fi
done

echo "✅ Emergency shutdown complete"

GCP Emergency Shutdown

#!/bin/bash
# gcp-emergency-shutdown.sh

echo "🚨 GCP Emergency Cost Control"
echo "============================"

read -p "⚠️ This will shut down non-essential resources. Continue? (y/N) " -n 1 -r
echo
if [[ ! $REPLY =~ ^[Yy]$ ]]; then
    echo "❌ Emergency shutdown cancelled"
    exit 1
fi

# Stop development Cloud SQL instances
echo "🛑 Stopping development Cloud SQL instances..."
gcloud sql instances list --filter="labels.environment=development" --format="value(name)" | while read instance; do
    if [ ! -z "$instance" ]; then
        echo "Stopping $instance..."
        gcloud sql instances patch $instance --activation-policy=NEVER
    fi
done

# Stop development Compute Engine instances
echo "🛑 Stopping development Compute Engine instances..."
gcloud compute instances list --filter="labels.environment=development AND status=RUNNING" --format="value(name,zone)" | while read instance zone; do
    if [ ! -z "$instance" ]; then
        echo "Stopping $instance in $zone..."
        gcloud compute instances stop $instance --zone=$zone
    fi
done

echo "✅ Emergency shutdown complete"

📈 Cost Reporting

Monthly Cost Report

#!/bin/bash
# monthly-cost-report.sh

echo "📊 Monthly Infrastructure Cost Report"
echo "===================================="

MONTH=$(date +%Y-%m)
LAST_MONTH=$(date -d "last month" +%Y-%m)

echo "Report for: $MONTH"
echo "Comparison with: $LAST_MONTH"
echo ""

if [ "$PROVIDER" = "aws" ]; then
    # AWS monthly report
    echo "💰 AWS Costs:"
    
    # Current month
    CURRENT_COST=$(aws ce get-cost-and-usage \
      --time-period Start=${MONTH}-01,End=$(date +%Y-%m-%d) \
      --granularity MONTHLY \
      --metrics BlendedCost \
      --query 'ResultsByTime[0].Total.BlendedCost.Amount' \
      --output text)
    
    echo "Current month (${MONTH}): \$${CURRENT_COST}"
    
    # Previous month
    LAST_MONTH_COST=$(aws ce get-cost-and-usage \
      --time-period Start=${LAST_MONTH}-01,End=${MONTH}-01 \
      --granularity MONTHLY \
      --metrics BlendedCost \
      --query 'ResultsByTime[0].Total.BlendedCost.Amount' \
      --output text)
    
    echo "Previous month (${LAST_MONTH}): \$${LAST_MONTH_COST}"
    
    # Calculate change
    if [ ! -z "$CURRENT_COST" ] && [ ! -z "$LAST_MONTH_COST" ]; then
        CHANGE=$(echo "$CURRENT_COST $LAST_MONTH_COST" | awk '{printf "%.2f", ($1-$2)/$2*100}')
        echo "Change: ${CHANGE}%"
    fi

elif [ "$PROVIDER" = "gcp" ]; then
    # GCP monthly report
    echo "💰 GCP Costs:"
    
    PROJECT_ID=$(gcloud config get-value project)
    
    # Current month costs (requires BigQuery billing export)
    bq query --use_legacy_sql=false \
    "SELECT
      ROUND(SUM(cost), 2) as current_month_cost
    FROM \`$PROJECT_ID.billing.gcp_billing_export_v1_*\`
    WHERE DATE(usage_start_time) >= '${MONTH}-01'
      AND DATE(usage_start_time) < DATE_ADD('${MONTH}-01', INTERVAL 1 MONTH)"
fi

echo ""
echo "📋 Cost Breakdown by Environment:"

# Environment-specific costs would go here
echo "Development: [Calculate from tags/labels]"
echo "Production: [Calculate from tags/labels]"

Cost Optimization Recommendations

#!/bin/bash
# cost-optimization-recommendations.sh

echo "💡 Cost Optimization Recommendations"
echo "===================================="

if [ "$PROVIDER" = "aws" ]; then
    echo "🔍 AWS Recommendations:"
    
    # Get AWS Cost Explorer recommendations
    aws ce get-rightsizing-recommendation \
      --service EC2-Instance \
      --query 'RightsizingRecommendations[?RightsizingType==`Terminate`].[ResourceId,CurrentInstance.InstanceType,RightsizingType]' \
      --output table
    
    # Reserved Instance recommendations
    aws ce get-reservation-purchase-recommendation \
      --service EC2-Instance \
      --query 'Recommendations[].RecommendationDetails.[InstanceDetails.EC2InstanceDetails.InstanceType,RecommendedNumberOfInstancesToPurchase]' \
      --output table

elif [ "$PROVIDER" = "gcp" ]; then
    echo "🔍 GCP Recommendations:"
    
    # Get GCP recommender suggestions
    gcloud recommender recommendations list \
      --project=$(gcloud config get-value project) \
      --recommender=google.compute.instance.MachineTypeRecommender \
      --location=us-central1 \
      --format="table(name,description,primaryImpact.costProjection.cost.units)"
fi

echo ""
echo "📊 General Recommendations:"
echo "1. Review unused resources monthly"
echo "2. Implement auto-scaling policies"
echo "3. Use spot/preemptible instances for non-critical workloads"
echo "4. Set up lifecycle policies for storage"
echo "5. Monitor and optimize database performance"

Last updated: $(date) Version: 1.0.0