3.2 AWS Storage Services
Key Takeaways
- Amazon S3 is object storage with 99.999999999% (11 nines) durability — the most commonly used storage service on AWS.
- Amazon EBS provides block storage volumes for EC2 instances — like a virtual hard drive that persists independently of the instance.
- Amazon EFS is a fully managed, elastic file system that can be mounted by multiple EC2 instances simultaneously.
- S3 storage classes range from Standard (frequent access) to Glacier Deep Archive (rarely accessed, cheapest) — choose based on access frequency.
- AWS Storage Gateway connects on-premises environments to cloud storage for hybrid workloads.
AWS Storage Services
The Three Types of Cloud Storage
| Type | Description | AWS Service | Analogy |
|---|---|---|---|
| Object Storage | Flat structure, store files as objects with metadata | Amazon S3 | Filing cabinet with labeled folders |
| Block Storage | Data stored in fixed-size blocks, like a virtual hard drive | Amazon EBS | Hard drive attached to your computer |
| File Storage | Hierarchical file system (directories and files) | Amazon EFS, FSx | Network shared drive |
Amazon S3 (Simple Storage Service)
Amazon S3 is object storage built to store and retrieve any amount of data from anywhere. It is one of the most important AWS services for the exam.
S3 Key Features
- Durability: 99.999999999% (11 nines) — designed to sustain the loss of data in 2 facilities simultaneously
- Availability: 99.99% for S3 Standard
- Scalability: Virtually unlimited storage
- Objects can be up to 5 TB in size
- Buckets are the top-level containers for objects
- Bucket names must be globally unique across ALL AWS accounts
S3 Storage Classes
| Storage Class | Access Pattern | Min Storage Duration | Retrieval Time | Cost |
|---|---|---|---|---|
| S3 Standard | Frequent access | None | Milliseconds | $$$$ |
| S3 Intelligent-Tiering | Changing/unknown access patterns | None | Milliseconds | $$$$ (auto-optimized) |
| S3 Standard-IA | Infrequent access (once/month) | 30 days | Milliseconds | $$$ |
| S3 One Zone-IA | Infrequent access, non-critical | 30 days | Milliseconds | $$ |
| S3 Glacier Instant Retrieval | Rarely accessed, instant retrieval needed | 90 days | Milliseconds | $$ |
| S3 Glacier Flexible Retrieval | Archive, retrieval in minutes to hours | 90 days | Minutes to hours | $ |
| S3 Glacier Deep Archive | Long-term archive, rarely accessed | 180 days | Up to 12 hours | $ (cheapest) |
On the Exam: Know the access patterns for each storage class. Frequent access = Standard. Unknown/changing = Intelligent-Tiering. Infrequent = IA. Archive = Glacier. Long-term archive = Deep Archive.
S3 Key Concepts
| Concept | Description |
|---|---|
| Versioning | Keep multiple versions of an object; protect against accidental deletion |
| Lifecycle Policies | Automatically transition objects between storage classes or delete them after a time |
| Replication | Cross-Region Replication (CRR) or Same-Region Replication (SRR) for compliance and disaster recovery |
| Encryption | Server-side (SSE-S3, SSE-KMS, SSE-C) or client-side encryption |
| Access Control | Bucket policies, ACLs, IAM policies, S3 Access Points |
| Static Website Hosting | Host a static website directly from an S3 bucket |
Amazon EBS (Elastic Block Store)
Amazon EBS provides persistent block storage volumes for EC2 instances. Think of EBS as a virtual hard drive that you attach to your virtual server.
EBS Key Facts
- EBS volumes persist independently of the EC2 instance lifecycle
- EBS volumes are AZ-specific — they can only be attached to instances in the same AZ
- You can take snapshots of EBS volumes (stored in S3) for backup and disaster recovery
- Snapshots are incremental — only changed blocks are saved
- EBS volumes can be encrypted using AWS KMS
EBS Volume Types
| Type | Category | Use Case |
|---|---|---|
| gp3 / gp2 | General Purpose SSD | Boot volumes, dev/test environments, low-latency apps |
| io2 / io1 | Provisioned IOPS SSD | High-performance databases, I/O-intensive workloads |
| st1 | Throughput Optimized HDD | Big data, data warehouses, log processing |
| sc1 | Cold HDD | Infrequent access, lowest-cost HDD |
On the Exam: EBS is for EC2 block storage. Key difference from S3: EBS = block storage (like a hard drive), S3 = object storage (like a file repository). EBS is AZ-specific; S3 is Region-wide.
Amazon EFS (Elastic File System)
Amazon EFS is a fully managed, scalable, elastic file system for Linux workloads.
| Feature | Detail |
|---|---|
| Protocol | NFS (Network File System) v4 |
| OS Support | Linux only (use FSx for Windows) |
| Scaling | Automatically grows and shrinks |
| Access | Can be mounted by multiple EC2 instances concurrently |
| Availability | Multi-AZ by default |
| Use Cases | Content management, web serving, shared file storage |
Amazon FSx
Amazon FSx provides fully managed third-party file systems:
| Service | File System | Use Case |
|---|---|---|
| FSx for Windows File Server | Windows NTFS | Windows applications needing SMB protocol |
| FSx for Lustre | Lustre | High-performance computing (HPC), ML training |
| FSx for NetApp ONTAP | NetApp | Enterprise workloads migrating from on-premises |
| FSx for OpenZFS | OpenZFS | Linux workloads requiring ZFS features |
Storage Comparison
| Feature | S3 | EBS | EFS |
|---|---|---|---|
| Type | Object | Block | File |
| Scope | Region | AZ | Region (Multi-AZ) |
| Access | API/HTTP | Single EC2 instance* | Multiple EC2 instances |
| Max Size | Unlimited (5 TB per object) | 64 TiB per volume | Petabytes |
| Durability | 11 nines | 99.999% | 99.999999999% |
| Pricing | Per GB stored + requests | Per GB provisioned | Per GB used |
*EBS Multi-Attach is available for io2 volumes but only within one AZ.
What level of durability does Amazon S3 Standard provide?
A company needs to archive data that is rarely accessed and can tolerate retrieval times of up to 12 hours. Which S3 storage class is MOST cost-effective?
What is the key difference between Amazon EBS and Amazon S3?
Which AWS storage service allows multiple EC2 instances to share a file system simultaneously?