Amazon Web Services has announced the release of Glacier, a low-cost storage service that has been customised for data archiving and backup, the company has said.
Areas that are a good fit for Glacier include media archives, financial and healthcare records, raw genomic sequence data, long-term database backups, and information that has to be retained for regulatory compliance, it said.
Today, enterprises typically over-pay for data archiving, because of the upfront cost of an archiving solution, and also because many over-provision to make sure they have enough capacity for data redundancy and unexpected growth, according to Amazon.
Storing data using Glacier costs from US$0.01 per gigabyte per month. Its existing Simple Storage Service (S3) in comparison costs from between $0.125 and $0.055 per gigabyte per month for standard storage. To get the lowest prices, data volumes above 5,000 terabyte are needed. Unlike S3, there is no free tier offered for Glacier.
The two storage services are complementary. Enterprises should use S3 if they need low latency or frequent access to their data, while Glacier is a better fit if low storage cost is paramount; data is rarely retrieved; and data retrieval times of several hours are okay, according to Amazon.
Also, to tie the two closer together, Amazon will introduce an option that allows users to move data between S3 and Glacier based on lifecycle policies in the coming months.
There is no limit to the amount of data users can store in Glacier, and they can also choose in which of Amazon’s regions the data is stored. To improve durability, data is synchronized across multiple facilities, Amazon said.
Data is stored in Glacier as archives, which can store up to 40TB. They can represent a single file or several files that are uploaded as a single archive. Archives can then be organised as vaults, the access to which can be controlled through Amazon’s Identity and Access Management (IAM) service, the company said.