Task Statement 4.1: Design cost-optimized storage solutions.
📘AWS Certified Solutions Architect – (SAA-C03)
1. Amazon S3 (Simple Storage Service)
What it is:
Amazon S3 is object storage. Think of it like a big online filing cabinet where you can store files (objects) such as documents, backups, images, or logs.
Key Features:
- Highly durable (11 nines of durability — 99.999999999%).
- Scalable: can store unlimited data.
- Can set storage classes for cost optimization.
Storage Classes (Cost Optimization Focus):
- S3 Standard – frequently accessed data. Higher cost.
- S3 Intelligent-Tiering – automatically moves objects between frequent and infrequent access tiers based on usage. Good for unknown access patterns.
- S3 Standard-Infrequent Access (IA) – for data you access occasionally. Lower storage cost, but a retrieval fee applies.
- S3 One Zone-IA – cheaper IA option stored in a single availability zone. Less resilient.
- S3 Glacier – long-term archival, slow retrieval (minutes to hours), very cheap.
- S3 Glacier Deep Archive – cheapest long-term storage, retrieval in hours.
Use Cases in IT:
- Storing backups, logs, or database snapshots.
- Storing static website assets (images, HTML files).
- Archiving old data that isn’t frequently accessed.
Cost Tips for the Exam:
- Use lifecycle policies to automatically move data from Standard → IA → Glacier.
- Use Requester Pays buckets if external users are accessing your data.
2. Amazon EBS (Elastic Block Store)
What it is:
Amazon EBS is block storage attached to EC2 instances. Think of it as a virtual hard drive for a server.
Key Features:
- You can attach/detach EBS volumes to EC2 instances.
- Supports snapshots for backup (stored in S3).
- Different volume types for cost and performance.
Volume Types (Cost and Performance):
- General Purpose SSD (gp3) – balanced performance and cost, default choice.
- Provisioned IOPS SSD (io2) – high-performance for databases, expensive.
- Throughput Optimized HDD (st1) – low-cost, good for large sequential workloads.
- Cold HDD (sc1) – cheapest, rarely accessed data.
Use Cases in IT:
- Storing database data for EC2-hosted databases.
- OS disks for EC2 instances.
- Log processing or file systems with frequent updates.
Cost Tips for the Exam:
- Choose gp3 over gp2 for general workloads (cheaper & flexible).
- Delete unattached volumes to avoid paying for unused storage.
- Use EBS snapshots for cost-efficient backups instead of keeping full volumes all the time.
3. Amazon EFS (Elastic File System)
What it is:
Amazon EFS is managed file storage for multiple EC2 instances. It’s like a shared folder that many servers can access at the same time.
Key Features:
- Scales automatically as you add/remove files.
- Can be accessed by multiple EC2 instances simultaneously.
- Two performance modes:
- General Purpose – for most workloads.
- Max I/O – for large-scale parallel workloads.
- Storage classes for cost:
- Standard – frequently accessed.
- Infrequent Access (EFS IA) – less accessed, cheaper.
Use Cases in IT:
- Shared storage for web servers (all servers need same files).
- Home directories for multiple users.
- Big data workloads where multiple instances process the same data.
Cost Tips for the Exam:
- Use EFS IA for files not accessed often.
- EFS can scale automatically, so avoid pre-provisioning.
- Lifecycle policies: move inactive files to IA to reduce cost.
4. Amazon FSx
What it is:
Amazon FSx provides fully managed Windows or Lustre file systems. It’s for workloads that need specific file system features like Windows SMB or high-performance parallel file systems.
Types:
- FSx for Windows File Server – SMB file system for Windows-based applications.
- FSx for Lustre – high-performance file system for compute-heavy workloads (like HPC or ML).
Use Cases in IT:
- FSx Windows: file shares for Windows applications or Active Directory integration.
- FSx Lustre: large data processing pipelines, HPC workloads.
Cost Tips for the Exam:
- Choose FSx type based on workload: Windows → application servers; Lustre → heavy compute/data processing.
- Avoid over-provisioning storage and throughput.
5. Exam Tips: How to Choose the Right Storage for Cost Optimization
| AWS Service | Access Pattern | Performance | Cost Optimization Strategy |
|---|---|---|---|
| S3 | Object storage, accessed via API | High durability, scalable | Use lifecycle policies, choose the right storage class (IA/Glacier) |
| EBS | Block storage attached to EC2 | Low-latency, high IOPS | Use gp3 for general use, delete unused volumes, snapshots for backups |
| EFS | Shared file system across multiple EC2 | Auto-scalable, POSIX-compliant | Use IA storage for infrequent files, lifecycle policies |
| FSx | Specialized file systems | Windows SMB or Lustre HPC | Choose correct type for workload, avoid over-provisioning |
Key Exam Takeaways:
- Object vs Block vs File: Know which service stores data in which way.
- Cost optimization often involves right-sizing, choosing proper storage class, and lifecycle policies.
- Multi-instance access → EFS or FSx; single EC2 access → EBS.
- Long-term archival → S3 Glacier/Deep Archive.
- Frequently changing data → EBS or S3 Standard.
