CLF-C02
Deep Dive
AWS offers multiple storage options for different use cases. The CLF-C02 tests your ability to choose the right service for the scenario: S3 for objects, EBS for EC2 volumes, EFS for shared file systems.
Contents
📦
Object storage
Amazon S3
Flat files, accessible via HTTP. No real folder structure (only prefixes). Scales infinitely.
💾
Block storage
Amazon EBS
A volume attached to an EC2 instance. Like a virtual hard drive. Formatted with a file system.
📁
File storage
Amazon EFS
A shared NFS file system. Multiple EC2 instances can mount the same EFS simultaneously.
Object storage with 11 nines of durability
99.999999999% durability = if you store 10 million objects, you'd expect to lose one every 10,000 years
Amazon S3 is object storage built to store and retrieve any amount of data from anywhere. Each object can be up to 5 TB in size. Objects are stored in buckets with globally unique names.
Key S3 features
S3 use cases
An audit client can download files without signing in. Find the 2 configuration problems and fix them — no guided steps.
+1 more...
The CLF-C02 exam tests when to use each class. General rule: cheaper to store = more expensive and/or slower to retrieve.
S3 Standard
Access: FrequentS3 Intelligent-Tiering
Access: Variable / unknownS3 Standard-IA
Access: InfrequentS3 One Zone-IA
Access: InfrequentS3 Glacier Instant Retrieval
Access: Archived (instant access when needed)S3 Glacier Flexible Retrieval
Access: Archived (retrieval in minutes or hours)S3 Glacier Deep Archive
Access: Deep archive (retrieval in 12-48h)No guided steps: read a scenario with several stacked requirements and pick, out of 4 S3 classes, the only one that meets all of them.
+1 more...
High-performance block storage for EC2 instances
Amazon EBS provides persistent block storage volumes for EC2 instances. The data persists independently of the EC2 instance's lifecycle. An EBS volume can only be attached to ONE EC2 instance at a time (with exceptions: EBS Multi-Attach for io1/io2).
| Type | Description | IOPS | Use cases |
|---|---|---|---|
| gp3 (SSD) | General purpose SSD. The recommended default. | Up to 16,000 | Boot volumes, web apps, dev/test databases |
| gp2 (SSD) | Previous-generation general purpose SSD (deprecated, use gp3). | Up to 16,000 | Legacy. gp3 is cheaper and more flexible. |
| io2 Block Express | High-durability provisioned IOPS SSD. | Up to 256,000 | Mission-critical databases (Oracle, SAP HANA), I/O-intensive workloads |
| io1 (SSD) | Provisioned IOPS SSD. High performance. | Up to 64,000 | Production transactional databases |
| st1 (HDD) | Throughput-optimized HDD. For frequently accessed data. | 500 MB/s throughput | Big data, data warehouses, log processing |
| sc1 (HDD) | Cold storage HDD. The most economical. | 250 MB/s throughput | Backups, infrequently accessed data |
A fully managed, shared NFS file system
Amazon EFS provides a simple, serverless, elastic NFS (Network File System) file system that can be shared across multiple EC2 instances simultaneously, even across different Availability Zones.
Advantages of EFS
Limitations
Instance Store is storage physically attached to the EC2 host (not networked storage like EBS). It offers very high I/O speed, but data is lost when the instance is stopped or terminated.
Characteristics
When to use Instance Store
Instance Store vs EBS
• Instance Store: faster, but ephemeral (data is lost)
• EBS: persists even if the instance restarts
• For important data → always use EBS (or S3)
Fully managed file systems for specific use cases
Amazon FSx provides specialized file systems for use cases EFS doesn't cover: Windows (SMB), HPC (Lustre), etc. Key takeaway for the exam: EFS = Linux/NFS. FSx = specific use cases.
FSx for Windows File Server
An SMB file system compatible with Windows and Active Directory. Ideal for Windows applications that need shared file storage. Supports DFS, NTFS, Active Directory.
FSx for Lustre
A high-performance file system for HPC, ML, and video processing. Designed for workloads requiring hundreds of GB/s of throughput and millions of IOPS.
FSx for NetApp ONTAP
Compatible with NetApp ONTAP. For migrations from on-premises NetApp environments to AWS without changing applications.
FSx for OpenZFS
A managed ZFS file system. Instant snapshots, volume cloning, and built-in compression.
S3 Lifecycle Policies
Automates the transition of S3 objects between storage classes based on age, reducing costs without manual intervention.
Example: access logs — active for 30 days, archived after 60 days, deleted after 1 year. Fully automated.
AWS Backup
A centralized backup service for multiple AWS services. Create backup policies (frequency, retention, transition to cold storage) that apply to EC2, EBS, RDS, DynamoDB, EFS, FSx, and more.
AWS Elastic Disaster Recovery
Continuously replicates servers from on-premises or the cloud (any source) into AWS. In the event of a disaster, enables recovery in minutes with an RPO of seconds and an RTO of minutes.
Current status: Snowcone was discontinued in November 2024, Snowmobile is retired, and Snowball Edge has not accepted new customers since November 7, 2025. For new customers, AWS recommends AWS DataSync (network transfer), AWS Data Transfer Terminal (physical drop-off at an AWS location), or AWS Outposts for edge compute. The CLF-C02 exam may still mention the Snow Family.
Physical devices for migrating large volumes of data into AWS
When you have petabytes of data and internet bandwidth isn't enough (it would take months/years), AWS ships a physical device to your datacenter. You copy the data to the device, ship it back, and AWS loads it into S3.
AWS Snowcone (discontinued)
Capacity: 8 TB (HDD) / 14 TB (SSD)
Weight: 2.1 kg
Migration of small volumes, edge computing in remote or space-constrained locations. Compatible with DataSync.
AWS Snowball Edge (no new customers)
Capacity: 80 TB (Storage Optimized) / 40 TB (Compute Optimized)
Weight: 22 kg
Migration of tens of TB. Also has compute capacity (EC2, Lambda on the device). For edge computing in locations without connectivity.
AWS Snowmobile (retired)
Capacity: Up to 100 PB
Weight: 45-foot shipping container on a truck
Exabyte-scale migrations. AWS drives a truck to your datacenter. For companies with tens of petabytes of historical data.
| Dimension | S3 | EBS | EFS |
|---|---|---|---|
| Type | Objects | Blocks (volume) | Files (NFS) |
| Access | REST API (HTTP/S) | Attached to EC2 (block device) | Mounted file system (NFS v4) |
| Concurrency | Millions of simultaneous clients | 1 instance (except Multi-Attach) | Thousands of simultaneous EC2 instances |
| OS compatibility | Any (HTTP) | Linux and Windows | Linux only (NFS) |
| Max object size | 5 TB per object, unlimited total | Up to 64 TB per volume | No limit (scales automatically) |
| Durability | 11 nines (multi-AZ) | 99.999% (specific AZ) | Multi-AZ (regional) |
| Use cases | Files, backups, static websites, data lakes | EC2 OS, databases, boot volumes | Shared code, CMS, multi-instance web servers |
| Approx. price | From $0.023/GB/month | From $0.08/GB/month (gp3) | From $0.30/GB/month |
Create your first object storage bucket and configure access permissions.
+1 more...
Did you get this topic?
Test what you just learned
A company has a web server with multiple EC2 instances in an Auto Scaling Group. They need every instance to share the same directory of configuration files and application assets. Which AWS storage service is the best fit?
Put what you learned to the test with original CLF-C02 exam questions.
Start CLF-C02 mock examUnlimited mock exams with the Pro plan
Log in to track your progress.