AWS

An overview of different storage classes on S3

Amazon’s simple storage service is an online object storage service provided by AWS. S3 stores data inside the buckets in the form of objects in multiple availability zones for high availability. S3 bucket names are universally unique, and the name of the new bucket should not exist in any other AWS account. AWS provides different storage classes to store the data on S3 depending upon the performance metrics like durability, availability, and cost. Following are the S3 storage classes provided by AWS.

  • S3 Standard
  • S3 Intelligent-Tiering
  • S3 Standard-IA (standard infrequently accessed)
  • S3 One Zone-IA (one zone infrequently accessed)
  • S3 Glacier
  • S3 Glacier Deep Archive

S3 Standard

S3 Standard is the standard S3 storage class for storing the data that is frequently accessed. It is the default storage class of S3 to store the objects. It provides 99.999999999% (11 9’s) durability and 99.99% of availability of objects stored in it. It stores data in multiple availability zones and provides lower first-byte latency and high throughput performance.

Key features

Following are some key features of the S3 Standard storage class.

  • Designed to provide 99.999999999% (11 9’s) durability of objects stored in it.
  • Designed for the availability of 99.99% over a year.
  • Stores data in multiple availability zones.
  • Provides first-byte latency in milliseconds and high throughput performance
  • Lifecycle management to automatically move objects to other storage classes over time.

Use cases

Following are some use cases for the S3 Standard storage class.

  • Cloud applications
  • Dynamic websites
  • Content distribution
  • Mobile and gaming applications
  • Big data analytics

S3 Intelligent-Tiering

S3 Intelligent-Tiering class is the S3 storage class that saves cost by moving the data into different storage classes depending upon the data access pattern. This storage class moves data into different classes without impacting the performance. The cost for this storage class depends upon the data access pattern. A small amount is also charged per month for monitoring the data access pattern.

Key features

Following are the key features of the S3 Intelligent-Tier.

  • Automatic cost optimization
  • Designed to provide 99.999999999% (11 9’s) durability of objects stored in it.
  • Designed for the availability of 99.9% over a year
  • Stores data in multiple availability zones
  • No limit on minimum storage duration

Use cases

Some of the use cases of the S3 Intelligent-Tier are listed below.

  • Data with random access patterns

S3 Standard-IA (standard infrequently accessed)

S3 Standard-IA is the storage class used to store the infrequently accessed data, requiring rapid data access and high availability. This storage class provides high throughput, low latency, and high availability of S3 Standard. It charges a low per GB storage price compared to S3 Standard, but a small amount is charged every time you retrieve data from this storage class. Due to the retrieval charges, this storage class may cost higher than the S3 Standard when data is frequently accessed.

Key features

Following are some key features of the S3 Standard-IA storage class.

  • Low latency and high throughput same as S3 Standard
  • Designed to provide 99.999999999% (11 9’s) durability of objects stored in it.
  • Designed for 99.99% availability
  • Stores data in multiple availability zones
  • Lifecycle management to automatically move objects to other storage classes over time.

Use cases

Following are the use cases of the S3 Standard-IA storage class.

  • Long term storage
  • Primary backups
  • Datastore for disaster recovery files

S3 One Zone-IA (one zone infrequently accessed)

S3 One Zone-IA is the storage class used to store infrequently accessed data and requires rapid access to the data. Unlike other S3 storage classes, S3 One Zone-IA does not store data in multiple availability zones, affecting its high availability. It charges a lower per GB storage price than S3 Standard and S3 Standard-IA, but a small amount is charged every time you retrieve data from this storage class. Due to the retrieval charges and single availability zone, this storage class can not be used for primary backups.

Key features

Following are some key features of the S3 One Zone-IA storage class.

  • Low latency and high throughput same as S3 Standard
  • Designed to provide 99.999999999% (11 9’s) durability of objects in a single availability zone
  • Designed for 99.5% availability
  • Lifecycle management to automatically move objects to other S3 storage classes

Use cases

Following are the use cases of the S3 One Zone-IA storage class.

  • Secondary backups
  • Storage for easily recreatable data
  • Storage for S3 replicated data from another AWS region

NOTE: As S3 One Zone-IA stores data in a single availability zone, data in this storage class will be lost in case of availability zone destruction.

S3 Glacier

S3 Glacier is a storage class for data archiving. It is a secure and low price storage class, and sometimes it can cost lower than on-premises solutions. S3 Glacier cost depends upon the data retrieval option, which varies from a few minutes to hours. This storage class is cheaper than other active S3 storage classes like (S3 Standard, S3 Intelligent-Tiering, S3 Standard-IA, S3 One Zone-IA).

Key features

Following are some important key features for the S3 Glacier storage class.

  • Designed to provide 99.999999999% (11 9’s) durability of objects stored in it.
  • Designed for 99.99% availability
  • Low-cost storage for long term data archival
  • Retrieval time configuration from minutes to hours
  • Data resilience across multiple availability zones

Use cases

Following are use cases for the S3 Glacier storage class.

  • Data Archive
  • Data bearing retrieval time from minutes to hours
  • Long term data storage

S3 Glacier Deep Archive

S3 Glacier Deep Archive is the lowest price S3 storage class used for long-term data storage that may be accessed somehow once a year. This storage class is useful for regulated industries that store data for about 5 to 10 years and access this data almost once a year. The retrieval time for the data stored in S3 Glacier deep archive is about 12 hours.

Key features

Key features of the S3 Glacier Deep Archive are listed below.

  • Designed to provide 99.999999999% (11 9’s) durability of objects stored in it.
  • Designed for 99.99% availability
  • Low-cost storage for long term data archival
  • Retrieval time configuration from 12 hours to 48 hours
  • Data resilience across multiple availability zones
  • Alternative to magnetic tape libraries

Use cases

Following are the use cases of the S3 Glacier Deep Archive storage class.

  • Data storage class for highly-regulated industries
  • Long term data storage
  • Data bearing retrieval time from 12 hours to 48 hours

Comparison between different storage classes

Following is the table that shows the characteristics of all the storage classes of the S3 service.

S3 Standard S3 Intelligent-Tiering S3 Standard-IA S3 One Zone-IA S3 Glacier S3 Glacier Deep Archive
Durability 99.999999999% (11 9’s) 99.999999999% (11 9’s) 99.999999999% (11 9’s) 99.999999999% (11 9’s) 99.999999999% (11 9’s) 99.999999999% (11 9’s)
Availability 99.99% 99.9% 99.9% 99.5% 99.99% 99.99%
Availability SLA 99.9% 99% 99% 99% 99.9% 99.9%
Availability Zones ≥3 ≥3 ≥3 1 ≥3 ≥3
Minimum capacity charge per object N/A N/A 128kb 128kb 40kb 40kb
Minimum storage duration charge N/A N/A 30 days 30 days 90 days 180 days
Retrieval charge N/A N/A per GB retrieved per GB retrieved per GB retrieved per GB retrieved
First-byte latency milliseconds milliseconds milliseconds milliseconds select minutes to hours select hours
Storage type Object Object Object Object Object Object
Lifecycle transition Yes Yes Yes Yes Yes Yes

Conclusion

AWS S3 provides different storage classes to store the data for different use cases. Each storage class has its own key features and use cases, and you can store your data in any class depending upon the nature of the data and access pattern. This guide gives a brief overview of all the storage classes provided by AWS and their use cases.

About the author

Zain Abideen

A DevOps Engineer with expertise in provisioning and managing servers on AWS and Software delivery lifecycle (SDLC) automation. I'm from Gujranwala, Pakistan and currently working as a DevOps engineer.