This content is part of the Essential Guide: AWS analytics tools help make sense of big data

How does EFS compare to other AWS storage options?

Amazon Elastic File System is one of a handful of choices for data storage. What are its strengths and weaknesses compared to other storage services?

Cloud storage options are proliferating to accommodate a growing range of user and workload needs. Amazon Elastic File System targets automatic scalability and multi-tenancy with high-performance solid-state disks.

Elastic File System (EFS) eliminates the need to deliberately provision storage and monitor capacity, though administrators must take greater care to review usage reports and implement policies that review and limit storage use. Otherwise it's easy to lose track of EFS use -- especially when multiple EC2 instances are using the storage -- and wind up with uncontrolled storage expenses.

Amazon EFS provides flexible solid-state disk (SSD)-based storage on demand for Amazon EC2 instances, scaling up and down automatically as storage needs change, shared by multiple EC2 instances.

EFS can be particularly beneficial for temporary, highly-scalable, performance-sensitive tasks like big data projects, development environments or other cloud initiatives, where storage needs are short term, substantial and unpredictable. However, EFS data can also be retained over the long term, serving as data repositories for media servers and other services. In these ways, EFS can be seen as one of the more beneficial AWS storage options.

Other AWS storage options, such as Amazon Simple Storage Service (S3), offer many of the same attributes found with EFS, including durability, high-availability, security and scalability. But S3 is not built on solid-state drives and isn't designed for the type of automatic usage-driven scalability and multi-tenancy provided by EFS. S3 is less expensive per GB than EFS, but S3 works on usage tiers and adds charges based on data transfer activities. S3 is often used for data repositories, backups and archives, disaster recovery, big data and cloud tasks.

Amazon Glacier offers low-cost, secure storage intended primarily for backup and archiving tasks. Glacier doesn't provide the performance of EFS or S3, but working data can easily be exchanged between Glacier's archival storage and S3 instances. More capacity can also be added as needs change. Glacier instances are often used for tasks like healthcare data retention, archiving for compliance, media storage and other large-scale, long-term storage needs.

Amazon Elastic Block Store (EBS) can be provisioned to provide long-term block storage volumes for EC2 instances. EBS volumes are associated with single EC2 instances, so there is no multi-tenancy, and the SSD-based storage is scalable through deliberate provisioning. Generally, EBS is used to provide backup, disaster recovery storage or more traditional storage for workloads in the cloud, such as relational databases and other enterprise applications.

Finally, Amazon is not the only provider to offer cloud-based storage. Google offers several versions of its cloud storage services including standard storage with high durability, availability and performance. Durable Reduced Availability Storage offers a low-cost alternative at the expense of occasional disruption. And Google Cloud Storage Nearline is intended as an economical alternative for archival storage tasks. Similarly, Microsoft Azure Storage promises high-capacity, high-performance, low-latency, storage scalable into the petabytes. Azure promotes this service for server message block file sharing across various cloud VMs.

Next Steps

AWS storage services and application requirements

Amazon Elastic File System roadmap details revealed

How to optimize memory for AWS database applications

Dig Deeper on Amazon S3 (Simple Storage Service) and backup