Cost-Optimized Architecture Patterns#

The last Domain 4 piece: not a tool, but a habit set. The exam’s “MOST cost-effective” questions are usually one of these patterns wearing a scenario costume.

The pattern checklist#

  1. Right-size - use Compute Optimizer; T-family burstables for spiky small work; don’t pay for a 16-vCPU box that idles at 5%
  2. Turn things off - dev/test schedules (Instance Scheduler), stop idle instances, delete unattached EBS (Elastic Block Store) volumes and idle ELBs (Elastic Load Balancing) - Trusted Advisor finds them
  3. Pay for commitment, not flexibility - steady state → Savings Plans/RI (Reserved Instance); see pricing-models
  4. Embrace Spot and serverless - batch/CI/stateless → Spot; event-driven, spiky → Lambda/Fargate (pay per use, no idle)
  5. Right storage class + lifecycle - s3-cost-and-lifecycle-policies
  6. Cache to cut backend load - CloudFront + ElastiCache = fewer DB instances, less egress (also better perf)
  7. Route data cheaply - VPC (Virtual Private Cloud) gateway endpoints (free S3 (Simple Storage Service) access vs NAT charges), keep inter-service traffic in-region/AZ (Availability Zone), CloudFront for internet egress
  8. S3 instead of compute - static hosting on S3/CloudFront beats an always-on web server

Cost-aware multi-tier design#

flowchart LR
    users[Users] --> cf[CloudFront - cached, cheap egress]
    cf --> alb[ALB]
    alb --> asg[ASG with Spot + On-Demand mix]
    asg --> sqs[SQS buffer]
    sqs --> lambda[Lambda consumers]
    asg & lambda --> ddb[DynamoDB on-demand]

Exam nuance: “most cost-effective” still must meet requirements. If the question says the workload is steady and latency-sensitive, Spot is wrong even though it’s cheapest. Cost answers are always “cheapest that works”.

Not: sinavda “dusuk maliyet” sorularinin cevaplari genelde su ucunden biridir: is yukunu kaldir (serverless), is yukunu kaydir (Spot/RI), veriyi ucuzlat (storage class/cache). Sadece “en dandik makineyi sec” cevabi neredeyse hic cikmaz.

Exam check#

Q1: A nightly batch job processes images for 2 hours and can restart from checkpoints. The MOST cost-effective compute is:

  • A) A reserved m5.2xlarge running 24/7
  • B) Spot instances via AWS Batch
  • C) An on-demand c5.4xlarge
  • D) A dedicated host
Cevabı göster

Answer: B. Short, restartable, fault-tolerant job = Spot (or Lambda if it fits - but “2 hours, checkpoints” points to Spot/Batch).

Q2: A small company needs to host a mostly static marketing site with a contact form, at the lowest cost. What should be used?

  • A) An EC2 (Elastic Compute Cloud) web server running 24/7
  • B) S3 static website hosting + CloudFront + Lambda/API Gateway for the form
  • C) A three-tier RDS-backed app
  • D) An ECS (Elastic Container Service) cluster
Cevabı göster

Answer: B. Static content doesn’t need always-on compute.

Ozet#

  • Right-size + kapati bos kaynaklari + taahhut ile indirim
  • Batch/test/stateless = Spot; spiky/olay-odakli = Lambda/Fargate
  • Cache (CloudFront/ElastiCache) = hem hiz hem maliyet
  • VPC gateway endpoint = NAT ucretlerini siler
  • Statik site = S3 + CloudFront, sunucu gerekmez
  • “En ucuz” daima “gereksinimleri karsilayan en ucuz” demektir
Backward cost-management-tools architecture-playbook Forward