Menu

AWS Announces S3 Intelligent Tiering, What is it Really?

AWS Delivers Automatic Cost Savings with S3 Intelligent Tiering

Amazon Simple Storage Service, better known as AWS S3, is one of the oldest and most widely used solutions from Amazon Web Services. S3 is designed to store terabytes of customer data with high availability and scale. Prior to AWS S3 Intelligent Tiering, Amazon cloud users had access to four storage class options:

  1. Standard – Designed for frequently accessed data.
  2. Standard-IA – Designed for long-lived, infrequently accessed data.
  3. One Zone-IA – Designed for long-lived, infrequently accessed, non-critical data.
  4. Glacier – Designed for long-lived, infrequent accessed, archived critical data.

The issue here is that access patterns can be different for the different objects. Though customers could build lifecycle policies to move the objects across storage classes, it was challenging to predefine the most appropriate storage class(s) because of the unpredictable nature of application adoption and usage. Even in scenarios where access frequency is known, customers may forget to make the proper class adjustments to optimize their budget.

Amazon observed this pain point and responded with the S3 Intelligent Tiering Storage Class – a smart solution to automatically manage the lifecycle of your S3 objects and help users to reduce overall AWS cloud cost. This innovation allows customers to focus on their applications while Amazon automatically addresses dynamic access patterns that often result in wasted budget if not monitored properly.

The new intelligent tiering incorporates two access tiers: frequent access and infrequent access. This moves objects that have not been accessed for 30 consecutive days to the infrequent access tier automatically. If the data is reaccessed after 30 days, the object will be moved back to the frequent access tier by the system. There are no associated retrieval fees, nor are there any fees for moving objects between tiers in the S3 Intelligent Tiering storage class.

Storage for objects in the frequent access tier is billed at the same rate as S3 Standard; storage for objects in the infrequent access tier is billed at the same rate as S3 Standard-Infrequent Access. Like in other storage classes you will be charged for storage, API calls, and data transfers. One notable limitation is that objects smaller than 128kb cannot be transitioned to the infrequent access tier hence they will be charged at the regular pricing.

Is AWS S3 Intelligent Tiering right for me?

It depends! This new S3 Intelligent Tiering Storage class will be a great fit for the customers when there is ambiguity around the access frequency of an S3 object. This functionality can be enabled while uploading new objects to S3. Customers also have an option of moving legacy objects to intelligent tiering via lifecycle policies. However, if customers are certain that their objects are accessed infrequently, then it is still recommended to use the Standard-IA storage class for better cloud cost saving opportunities. Otherwise, by allowing AWS to decide the appropriate tiers for each object and transition them as required, customers can not only improve cloud savings but also reallocate their time, which is arguably their most valuable.