Amazon Web Services has added dedicated instances to its virtual private cloud, allowing individual, physical servers to be rented from the Amazon public cloud.
The dedicated instances, announced on Monday, are designed for customers who have regulatory circumstances or security policies that require their data to be stored on physically isolated hardware, rather than on logically isolated virtual machines that share the same server.
"Each Virtual Private Cloud (VPC) and each EC2 instance running in a VPC now has an associated tenancy attribute. Leaving the attribute set to the value 'default' specifies the existing behaviour: a single physical machine may run instances launched by several different AWS customers," evangelist Jeff Barr wrote on the Amazon Web Services (AWS) blog on Monday. "Setting the tenancy of a VPC to "dedicated" when the VPC is created will ensure that all instances launched in the VPC will run on single-tenant hardware."
Dedicated instances can only be run from within AWS virtual private clouds.
Aside from a premium added to per-hour pricing for the specific instances used as dedicated machines, AWS will also levy an additional $10 (£6.26) per hour for every region across its cloud on which customer runs a dedicated instance on. AWS says this is because when a dedicated instance is run, it blocks out a server, which means AWS cannot sell that server's excess capacity to other customers.
The $10 charge per hour will take up proportionally less of the running costs of the instances, if customers launch more of them, AWS noted.
There is no guarantee that if one customer launches multiple dedicated instances they will share the same underlying hardware, Barr wrote. In fact, AWS will "go to some trouble to spread them out across several machines in order to minimise the effects of a hardware failure".
In March, AWS also announced a multitude of networking features for VPCs, which allows greater control over network topology.