Being Clear on the Virtual Private Cloud

Share: Print

The National Institute of Standards and Technology defines a hybrid cloud as a “composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability.”

That definition raises a key question: what does “bound together” mean? The unsatisfying answer: “It depends.”

An increasingly popular form of hybrid cloud, commonly called a “virtual private” cloud, has certain infrastructure elements, such as servers and storage, dedicated to each customer, while other infrastructure elements are multi-tenant (shared). This architecture addresses customers’ concerns about processing and data storage security, while still maintaining some resource sharing.

The mix of what’s dedicated to one customer versus what’s shared varies across vendors. From a sourcing perspective, it’s therefore imperative as a customer to understand the specific components and processes of a provider’s VPC solution, and whether they address the organization’s requirements for security, data privacy, redundancy, disaster recovery, and service quality.

For many VPC offerings, CPUs and storage are dedicated to each customer. Beyond these basics, however, little can be taken for granted, and customers must understand the technical implementations of how services are delivered.  For example, a solution that utilizes shared network links could be a problem, as the compliance requirements of some organizations preclude the sharing of any network infrastructure.

Pricing of VPC services also varies widely, creating further potential for confusion.  Some vendors offer usage-based models similar to public cloud pricing structures, e.g., charges for CPU utilization and disk storage.  Other pricing structures charge for the servers and storage dedicated to the customer so that, at least on the surface, pricing is similar to that of a traditional hosting environment. The latter approach is still called a “cloud” offering because it’s virtual and has pooled resources, broad network access, and self-service rapid provisioning – characteristics not typically found in a traditional infrastructure environment.

By taking advantage of the economies of scale of a cloud-based infrastructure, but in a way that is more secure than a fully public service, the hybrid/virtual private cloud model offers significant potential benefits. At present, however, the lack of standards and consistency for how vendors deliver and price VPC services makes it imperative for client organizations to have a precise understanding of the solution they are buying, as well as contractual provisions to address key considerations.


 
Share: