Amazon has unveiled a new cloud-based storage solution called Glacier that’s is designed to be used for storing large volumes of archived data, such old legal documents, raw footage for a video project, or large volumes of research data. The service is quite similar to S3, and offers the same the same 99.999999999% durability. However it is highly optimised, both in terms of pricing structure and the way the service works, for data that is accessed infrequently. For one, while S3 can serve as the backbone for serving content for web hosting solutions, or other data that has to be instantly accessed, and possibly modified, Glacier would not work in that capacity. To retrieve data from Glacier at any time, you need to initiate a job for it, a process that can take from 3 to 5 hours. The pricing of the service is very low compared to S3, and competing solutions. Amazon Glacier costs just $0.01 per GB per month, compared to base rate of $0.125 per GB per month for the Amazon S3 (or $0.093 for reduced redundancy storage). Additionally, to penalize frequent retrieval—or more accurately to promote infrequent retrieval—outgoing data from the service is free up to 5% of the average monthly storage. Those who transfer more than this will
have to pay transfer charges up to $0.01 per GB. As for Amazon S3, Amazon’s AWS Import/Export service is also available for Glacier. This service allows one to ship a large sized hard drive to Amazon to have that data uploaded quicker—useful if one has terabytes of data to upload. You can find out more about Glacier and its APIs from its page at the Amazon Web Services site.
