Prices for Microsoft's Azure backup service will fall by as much as two-thirds of current costs from next month as the result of a change to the way the company structures its charges.
From April 1, Microsoft will charge customers based on the size of each protected instance, in increments of 500 gigabytes.
Currently, backup is costed based on the total amount of data stored. Australian customers of Azure backup services are currently charged $0.20 per GB per month for more than 5GB of compressed data stored per month.
As of next month, Microsoft will reduce backup pricing for servers with larger amounts of data, adding a mix of fixed and per gigabyte charging.
From April, a small instance up to 50GB will cost around $5 per month plus $0.02 per GB; a medium instance up to 500GB will cost around $10 per month plus $0.02 per GB; and a large instance will cost $10 per each 500GB block and $0.02 per GB.
This means storing a 500GB backup in Azure will cost $30 per month from April, compared to the $100 that Australian customers are charged currently.
Microsoft said customers storing data for longer periods of time were being punished under the previous pricing model.
"The model for charging customers for every GB of backup data stored is an effective model when backup data is retained for short periods of time," Microsoft said.
"With the continual enhancements to the long term retention capabilities in Azure Backup and the ever increasing volumes of data being backed up, the per-GB model of charging becomes cost prohibitive for customers retaining data for long periods of time."
All existing backup customers will be transitioned to the new pricing model as of April 1.
Replication battle with AWS
Azure customers will also now have the flexibility to choose between Locally Redundant Storage (LRS) or GeoRedundant Storage Block Blob Storage (GRS).
LRS keeps three replicas of customer data within a single facility or region for protection, and GRS, which is the default zone storage option, replicates the data across two geographical regions. As with LRS, GRS keeps three copies of the data with each set kept in two regions.
Major Microsoft Azure rival Amazon Web Services today also announced new cross-replication functionality for its S3 backup service.
AWS said customers had told the company that regulatory requirements often required them to keep copies of critical data in locations hundreds of miles apart, which was not possible under the current model.
As such, AWS opted to launch a new 'cross-region replication feature' to address demand for geographically dispersed replication.
"Once enabled, every object uploaded to a particular S3 bucket is automatically replicated to a designated destination bucket located in a different AWS region," the company wrote in a post.
The feature is built on top of S3's existing versioning facility, AWS said. From there, customers choose the destination region and bucket and set up an identity and access management role.
This cross-region replication is available now. Users will pay the standard AWS prices for data transfer between region on top of additional data storage charges for the data in the destination bucket.