What is a private cloud?

Monday, April 27, 2015

Everything you hear in IT today mentions something about a cloud.  There is the Internet of Things that is stored somewhere in the cloud.  We hear about Platform as a Service (PaaS) or Infrastructure as a Service (IaaS).  There are public clouds and hybrid clouds and private clouds.  It is mentioned so often that it leads to misconceptions about what a cloud really means.   So what makes up cloud computing?

According to the National Institute of Standards and Technology (NIST), a cloud consists of 5 characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service.  

Clouds must allow consumers (in enterprises, the end users in the business units) to automatically provision resources without the need for manual steps or the intervention of an administrator.  This could include use of an application or computing or storage.  This self-service portal is the front-end view the consumer has for all of the automated resource delivery built on the back-end.

The cloud services are available over an existing network using standard mechanisms.   In other words, the services are easy to get to and not different from what the consumer expects.  It should support heterogeneous access through multiple browsers and clients (thick and thin).

Clouds are based on a multi-tenant model that allows multiple consumers to share resources.   The actual location of the resources is unknown to the consumer.  Resources may automatically move between different parts of the infrastructure as long the security, access, and performance meets negotiated Service Level Agreements (SLAs).

Capabilities of the cloud need to allow for both scale-up and scale-out.  Resources need to automatically grow and shrink to meet the needs of the consumer.   Consumers are no longer stuck with a one size fits all type approach.  Resources appear to be unlimited to the consumer and can be added at any time.  

Last, but certainly not least, clouds need to include a metering component that can measure the usage of the cloud by consumer.   Cloud computing falls into a utility model where consumers can sign up for resources and are billed on the resources they utilize.   The metering solution needs to measure the automatic elasticity and on-demand self-service features of the cloud.

Clouds can follow several service models:  Software as a Service (Saas), Platform as a Service(PaaS) and Infrastructure as a Service (IaaS).   It could take multiple blog posts to effectively describe each of these service models.  In the most fundamental of terms, here is a basic description of each.  SaaS offerings provide remote access to an application that completely utilizes the provider’s resources.  The PaaS model allows the customer to use their own applications on an underlying platform supplied by the provider.   IaaS providers supply infrastructure including CPU, network, storage and other resources that allow clients to run their own operating systems and applications.

A private cloud implementation takes the cloud characteristics defined by NIST and applies them for use by a single enterprise organization.  The primary reasons many enterprise organizations are creating private clouds is to enjoy the benefits of speed and flexibility to provision new applications along with increased customer satisfaction without sacrificing the control of the infrastructure.

Please visit the NIST’s definition of Cloud Computing: http://csrc.nist.gov/publications/nistpubs/800-145/SP800-145.pdf

APTARE—the global leader in IT Analytics—is the only company delivering unified backup and storage insights for heterogeneous IT environments across on-prem and hybrid-cloud