Cloud computing has brought efficiencies to businesses providing fast computational capacity and data storage with lower set up costs and ongoing expense. However, one of the largest discouragements from using cloud services, particularly public cloud services, comes from the inherent and unknown danger stemming from a shared platform—namely, the hypervisor.
A hypervisor acts as the central manager of all end user’s resources on a cloud network. These resources usually are embodied as virtual machines (VM), which act as an implementation of a computer and can run and execute programs as though they were running on physical hardware. A cloud computing facility can have hundreds of hypervisors that run thousands of virtual machines for many different end users.
Problems arise when a hacker exploits a vulnerability on a low-security VM running funny cat videos and from there infiltrates the hypervisor. At that point the hacker has access to all VMs under that hypervisor – potentially one allocated to a global bank or another high-value target. Thus, users with significant information assets could be discouraged from using the cloud since the potential loss associated with a successful attack is too large compared to the cost savings from switching to cloud resources.
To alleviate some of these cloud computing security concerns scientists have developed a VM allocation model for grouping VM of similar loss potential on the same hypervisor and creating hypervisor environments of similar total loss. When VMs of similar loss potential are grouped onto the same hypervisor, the effect of interdependency wherein the security of one user affects the security of another is greatly reduced.
- Reduces to the potential of grouping low security computing resources with high security resources in a cloud environment
- Setting up groups based on risk profile and differing levels of security allows for tiered pricing of cloud services
- US patent 9,832,220 available for license
- Potential for collaboration with Air Force researchers