Amazon Intros Intelligent Tiering For S3 Storage To Optimize Cost

Amazon on Monday unveiled a new automated cloud storage tiering capability for handling files that have not been accessed for 30 days.

Termed S3 Intelligent-Tiering, it allows users to get the benefit of lower-cost long-term storage without having an in-depth understanding of the data's access patterns, wrote Jeff Barr, chief evangelist for Amazon Web Services in a blog post.

The new tiering capability was unveiled at the start of the AWS re:Invent conference being held this week in Las Vegas.

[Related: Cloud Cornerstones: 19 New Ways To Connect Storage To AWS]

Until Monday, AWS offered four classes of service: Standard for frequently accessed data; Standard-IA for long-lived, infrequently accessed data; One Zone-IA for long-lived, infrequently accessed non-critical data; and Glacier for long-lived, infrequent accessed, archived critical data.

S3 Intelligent-Tiering is a new fifth class of service that automates movement of data between two access tiers, called frequent access and infrequent access, with the same low latency as the Standard storage class, Barr wrote.

If users want to use the S3 Intelligent-Tiering class, they can specify the class when adding new objects to S3. They can also use a Lifecycle Policy to transition other data to S3 Intelligent-Tiering, Barr wrote.

Once in S3 Intelligent-Tiering, storage of the objects in the frequent access tier is the same as if they were stored in the S3 Standard class, while objects in the infrequent access tier are stored at the same cost as if in the S3 Standard-IA (infrequent access) class.

Data in the S3 Intelligent-Tiering frequent access tier which is not accessed for 30 days is automatically moved to the infrequent access tier to take advantage of the lower cost, but is moved back to the frequent access tier if it is later accessed, Barr wrote.

"The bottom line: You save money even under changing access patterns, with no performance impact, no operational overhead, and no retrieval fees," he wrote.

Barr wrote that there are a number of considerations to keep in mind when deciding whether to use the S3 Intelligent-Tiering class. For data which customers are sure will be accessed infrequently, the Storage-IA class is better for cost reasons, while the S3 Intelligent-Tiering class is better for data which may see changes in access patterns.

Objects smaller than 128 KBs in size will never be transferred to the infrequent access tier, and objects that are kept in S3 for under 30 days will still be billed for a minimum of 30 days, he wrote.

Data stored in the Intelligent-Tiering class frequent access tier is billed the same rate as S3 Standard, while data on the infrequent access tier is billed the same as S3 Standard-IA.

"When you use Intelligent-Tiering, you pay a small monthly per-object fee for monitoring and automation; this means that the storage class becomes even more economical as object sizes grow. … S3 Intelligent-Tiering will automatically move data back to the frequent access tier based on access patterns but there is no retrieval charge," Barr wrote.

Solution providers who work with AWS said the new S3 Intelligent-Tiering service is a winner for customers and channel partners.

Amazon's initial unveiling of the Standard-IA class three years ago was already cool, said Bob Swanson, vice president of sales at dcVast, a Downers Grove, Ill.-based solution provider and AWS channel partner.

"We could have our cake and eat it too," Swanson told CRN. "It brought the governance and auto-tiering capabilities of other storage vendors into the cloud."

S3 Intelligent-Tiering takes this one step farther, Swanson said. Prior to this new class, it was important that clients categorized their data properly before moving it to the Standard-IA class. "If not, their costs could be higher, more than the S2 Standard tier if they access the data a lot," he said.

While data moved to the S3 Intelligent-Tiering infrequent access tier is subject to a "small monthly monitoring and automation fee" per object, the fact that there is no additional retrieval fee for data in the tier is a big cost savings for customers, Swanson said.

"This buffers the customer from getting jammed with crazy retrieval fees for data in S3-IA," he said.

S3 Intelligent-Tiering is a logical and highly asked-for feature set, said Brendan Walsh, vice president of 1901 Group, a Reston, Va.-based solution provider and AWS channel partner, via text message to CRN.

"The financial flexibility is an obvious value but needs to be carefully thought through for contractual and ITSEC (Information Technology Security Evaluation Criteria) policy perspectives, especially with public sector customers or highly regulated markets," Walsh wrote.

S3 Intelligent-Tiering is long overdue, said Sam Barhoumeh, founder and CEO of ReadyNetworks, an Evanston, Ill.-based solution provider and AWS channel partner.

"Everyone in the channel and on the consumer side will be excited about it, especially in the enterprise space where people need far more than just simple storage," Barhoumeh told CRN. "The features look at the data set, and says if the data is not touched for 30 days, it will be moved to a different container. It's still available, but customers pay less to store it."

For partners, the new class of storage can help create additional value for their clients, Barhoumeh said.

"Partners can create value by speaking with clients and looking at what their tenants look like to create templates based on performance issue so that if data is called on, there are no surprises," he said. "And this lets customers understand their infrastructure-as-a-service and what they need to do to prepare their data strategies."

S3 Intelligent-Tiering is a quality vs quantity play, Barhoumeh said.

"You now are focusing on the workload, understanding what customers need before throwing more cloud at it," he said. "You want to increase performance while reducing consumption costs."

However, Barhoumeh said, the biggest beneficiary of S3 Intelligent-Tiering may be Amazon itself.

"This allows Amazon to free up more valuable storage for data that needs it, saving them money that might otherwise be spent on having to build premium storage for people who do not want to pay for it," he said.