Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage News AWS Establishes Per-Second Billing for EC2 Instances

AWS Establishes Per-Second Billing for EC2 Instances

AWS instituted a per-second billing model for EC2 instances and EBS volumes on the 2nd of October. A few other AWS services like Amazon EMR, AWS Batch, and Elastic GPUs will have the same model with a one-minute minimum.

As Trevor Sullivan, AWS Solution Architect, explains on his blog:

Since 2006, when EC2 was launched, EC2 Instances were elastically billed by the hour. Now, that has all changed, and AWS customers are given even more granularity with their purchasing options.

Amazon bills customers per account or accounts, for usage by metering hourly for EC2 instances in different ways of payment; on demand, spot, reserved or dedicated pricing. The spot and reserved pricing both provide ways for cutting costs of using on-demand instances. The spot price for unused EC2 capacity can be 90% off the on-demand prices. And reserved capacity can lead to up 72% off based on a one-to-three-year term. The dedicated pricing for capacity offers customers 70% off when a reservation is made up front.

Microsoft offers a model predominantly on per-minute basis or unit of measure such as RU/s in Cosmos DB. Google on the other hand offer per-second billing for compute instances. The three cloud vendors have been competing with each other for years now regarding pricing. And it will be interesting how Microsoft and Google will follow up to this new pricing model.

There are however some subtle differences with regards to Amazon’s pricing of compute time. The per second billing only applies to Linux VMs, not Windows VMs, while Google’s and Microsoft’s prices cover all VMs whether they run Windows, Red Hat Linux or SUSE Linux.  Furthermore, Amazon still charges for other services and products in the AWS marketplace on an hourly basis. As Sullivan states:

List prices and Spot Market prices are still listed on a per-hour basis, but bills are calculated down to the second, as is Reserved Instance usage (you can launch, use, and terminate multiple instances within an hour and get the Reserved Instance Benefit for all of the instances). Also, bills will show times in the decimal form.

The metering changes to per second with a minimum of 60 seconds, which will be reflected on the bill with decimals in the hours.

Image Source:

The per-second billing brings consumers of cloud services closer to “pay as you go” or “only pay what you use”, and further pushes the commoditization of cloud services.

Rate this Article