2.2.1. Evaluating Storage Options (S3, EBS, EFS, Glacier, FSx, Storage Gateway)

💡 First Principle: Effective storage design hinges on classifying data by its access patterns, durability needs, and consistency requirements, then selecting the purpose-built storage service that optimizes performance and cost.

Scenario: You are designing the data persistence layer for a new enterprise application that will handle various types of data: frequently accessed user files, large archives for compliance, and high-performance block storage for a database. You need to select the most appropriate AWS storage services for each data type.

AWS offers a comprehensive suite of storage services, each tailored for specific use cases and access patterns.

• "Amazon S3 (Simple Storage Service)": An object storage service. Highly durable (11 nines), scalable, and cost-effective. Ideal for static content, backups, data lakes, disaster recovery, and serverless applications. Access via HTTP/S.
• "Amazon EBS (Elastic Block Store)": Block storage volumes for "EC2 instances". Provides persistent block-level storage volumes for use with "EC2". Ideal for database workloads, boot volumes, and applications requiring low-latency, raw block access. Attaches to a single "EC2 instance" in an "AZ".
• "Amazon EFS (Elastic File System)": A scalable, elastic, cloud-native NFS file system. Shared file storage (NFSv4) for "EC2 instances" and on-premises servers. Scalable, elastic, and highly available. Ideal for shared file systems, content management, development environments, and container workloads requiring shared storage.
• "Amazon Glacier"/"Glacier Deep Archive": Archive storage classes within "S3". Extremely low-cost, high-durability storage for infrequently accessed data with flexible retrieval times (minutes to hours). Ideal for long-term archiving and regulatory compliance.
• "Amazon FSx": Managed file systems for "Windows File Server", "Lustre", "NetApp ONTAP", and "OpenZFS". Provides high-performance, fully managed third-party file systems. Ideal for Windows-based applications, HPC workloads, and specific enterprise file sharing needs.
• "AWS Storage Gateway": A hybrid cloud storage solution. Connects on-premises environments to cloud storage. Provides local caching for frequently accessed data while leveraging AWS storage services ("S3", "EBS", "Glacier").

Visual: Storage Service Characteristics

Loading diagram...

⚠️ Common Pitfall: Choosing a storage service based on a familiar on-premises paradigm. For example, trying to use a file system ("EFS"/"FSx") for storing massive amounts of unstructured data like images or videos, when object storage ("S3") is the more scalable and cost-effective cloud-native solution.

Key Trade-Offs:

• Access Method vs. Performance: Object storage ("S3") offers massive scalability and durability but has higher latency for small, random access compared to block storage ("EBS"). File storage ("EFS") offers shared access but may not have the raw "IOPS" of a dedicated "EBS volume".

Reflection Question: Given the needs for frequently accessed user files, large archives for compliance, and high-performance block storage for a database, how would you select and justify the most appropriate AWS storage services (e.g., "S3", "Glacier", "EBS", "EFS") for each specific data type to optimize performance and cost?