S3 및 데이터 전송 비용 최적화

S3 스토리지 비용의 기본 원리

# S3 storage class costs (us-east-1 approximate):
# S3 Standard:              $0.023/GB/month
# S3 Intelligent-Tiering:   $0.023/GB (frequent) + $0.0125 (infrequent)
# S3 Standard-IA:           $0.0125/GB/month + retrieval fee
# S3 One Zone-IA:           $0.01/GB/month (single AZ)
# S3 Glacier Instant:       $0.004/GB/month + retrieval fee
# S3 Glacier Flexible:      $0.0036/GB/month + retrieval time
# S3 Glacier Deep Archive:  $0.00099/GB/month + retrieval time

S3 Intelligent-Tiering

# Enable Intelligent-Tiering on objects
aws s3api put-object \
  --bucket my-bucket \
  --key data/report.parquet \
  --body report.parquet \
  --storage-class INTELLIGENT_TIERING

# Or set as default for entire bucket
aws s3api put-bucket-intelligent-tiering-configuration \
  --bucket my-bucket \
  --id whole-bucket-tiering \
  --intelligent-tiering-configuration '{
    "Id": "whole-bucket-tiering",
    "Status": "Enabled",
    "Tierings": [
      {"AccessTier": "ARCHIVE_ACCESS", "Days": 90},
      {"AccessTier": "DEEP_ARCHIVE_ACCESS", "Days": 180}
    ]
  }'

자동 계층화를 위한 S3 Lifecycle 규칙

# Lifecycle rule: transition logs to archive
aws s3api put-bucket-lifecycle-configuration \
  --bucket app-logs-bucket \
  --lifecycle-configuration '{
    "Rules": [{
      "ID": "log-archive",
      "Filter": {"Prefix": "logs/"},
      "Status": "Enabled",
      "Transitions": [
        {"Days": 30,  "StorageClass": "STANDARD_IA"},
        {"Days": 90,  "StorageClass": "GLACIER_IR"},
        {"Days": 365, "StorageClass": "DEEP_ARCHIVE"}
      ],
      "Expiration": {"Days": 2555}
    }]
  }'

S3 Select 및 Glacier Select

# S3 Select: query a CSV object using SQL expression
aws s3api select-object-content \
  --bucket analytics-data \
  --key sales/2026-06.csv \
  --expression-type SQL \
  --expression "SELECT * FROM S3Object WHERE region = 'us-east'" \
  --input-serialization 'CSV={FileHeaderInfo=USE}' \
  --output-serialization 'CSV={}' \
  output.csv

# Only the matching rows are returned
# Saves data transfer cost vs downloading entire file

S3 Multipart Upload 및 Transfer Acceleration

# Multipart upload example
aws s3api create-multipart-upload \
  --bucket my-bucket --key large-file.zip

# Abort incomplete multipart uploads via lifecycle
aws s3api put-bucket-lifecycle-configuration \
  --bucket my-bucket \
  --lifecycle-configuration '{
    "Rules": [{
      "ID": "cleanup-incomplete-uploads",
      "Status": "Enabled",
      "AbortIncompleteMultipartUpload": {"DaysAfterInitiation": 7}
    }]
  }'

데이터 전송 비용 이해하기

# Data transfer cost examples (us-east-1):
# Internet (outbound):       $0.09/GB (first 10 TB)
# CloudFront -> internet:    $0.0085/GB (much cheaper!)
# Cross-region:              $0.02/GB
# Cross-AZ (same region):    $0.01/GB each direction
# Same AZ:                   $0.00/GB (free)
# S3 -> CloudFront:          $0.00/GB (free!)

# Cost reduction tip:
# S3 -> CloudFront -> users
# vs
# EC2 -> internet -> users
# CloudFront is 10x cheaper for data egress

데이터 전송 비용 절감을 위한 CloudFront

# Cost comparison for 100 TB/month of outbound traffic:
# Direct from EC2/S3 to internet:
#   100,000 GB x $0.09/GB = $9,000/month

# Via CloudFront (assuming 80% cache hit rate):
#   20,000 GB origin pull: free (S3->CF) or $0.02 (EC2->CF)
#   100,000 GB CF->internet: 100,000 x $0.0085 = $850
#   Total: ~$1,250 vs $9,000 = 86% savings

# Create CloudFront distribution with S3 origin
aws cloudfront create-distribution \
  --distribution-config file://cf-config.json

NAT Gateway 비용 방지를 위한 VPC 엔드포인트

# Create VPC Gateway Endpoint for S3 (free!)
aws ec2 create-vpc-endpoint \
  --vpc-id vpc-12345 \
  --service-name com.amazonaws.us-east-1.s3 \
  --vpc-endpoint-type Gateway \
  --route-table-ids rtb-private-AZ1 rtb-private-AZ2

# Traffic from EC2 in private subnet to S3:
# WITHOUT endpoint: EC2 -> NAT GW ($0.045/GB) -> S3
# WITH endpoint:    EC2 -> S3 via VPC endpoint ($0.00/GB)

# For DynamoDB:
aws ec2 create-vpc-endpoint \
  --service-name com.amazonaws.us-east-1.dynamodb \
  --vpc-endpoint-type Gateway

S3 요청 비용 최적화

# S3 Batch Operations: process millions of objects at once
aws s3control create-job \
  --account-id 123456789012 \
  --manifest '{
    "Spec":{"Format":"S3BatchOperations_CSV_20180820","Fields":["Bucket","Key"]},
    "Location":{"ObjectArn":"arn:aws:s3:::manifest-bucket/manifest.csv","ETag":"abc123"}
  }' \
  --operation '{
    "S3CopyObject": {
      "TargetResource": "arn:aws:s3:::destination-bucket",
      "StorageClass": "GLACIER"
    }
  }' \
  --report '{"Bucket":"arn:aws:s3:::report-bucket","Enabled":true}'

데이터 전송을 위한 AWS DataSync

# Create DataSync task to copy S3 bucket to another region
aws datasync create-task \
  --source-location-arn arn:aws:datasync:us-east-1:123:location/loc-source \
  --destination-location-arn arn:aws:datasync:us-west-2:123:location/loc-dest \
  --name 'DR-replication' \
  --options '{
    "VerifyMode": "ONLY_FILES_TRANSFERRED",
    "OverwriteMode": "ALWAYS",
    "TransferMode": "CHANGED"
  }'

# DataSync pricing: $0.0125/GB transferred

S3 가시성을 위한 Storage Lens

# Enable S3 Storage Lens dashboard
aws s3control put-storage-lens-configuration \
  --account-id 123456789012 \
  --config-id default \
  --storage-lens-configuration '{
    "Id": "default",
    "IsEnabled": true,
    "DataExport": {
      "S3BucketDestination": {
        "AccountId": "123456789012",
        "Arn": "arn:aws:s3:::my-storage-lens-export",
        "Format": "CSV",
        "OutputSchemaVersion": "V_1"
      }
    }
  }'

빠른 확인

강의 요약