Understanding
 the cloud options

There is no question that cloud computing and cloud services has seen breathtakingly rapid adoption and development over the last few years. This has culminated in widespread acceptance of the cloud as a valid part of modern IT deployments, which in turn puts the pressure on IT managers and CIOs about their enterprise cloud strategy.

On the flip side, misconceptions about the cloud continue to abound, courtesy of loosely used jargons from its early days. As the cloud becomes part and parcel of mainstream IT, it is important that marketing speak be replaced with more precise terminologies and a deeper understanding of what they mean. We examine two common cloud deployment options and they really mean to enterprises.

Multitenancy and virtualization

At its root, multitenancy refers to a software architecture that sees a single instance of a software serving multiple users, or tenants. Considered an important feature of cloud computing, multitenancy segregates tenants by giving each instance its own set of data, configuration and user management.

To be clear, cloud services can have differing degrees of multitenancy, depending on their core design and how their programming code is written. They also span all three layers of the cloud, namely infrastructure-as-a-service (IaaS), platform-as-a-service (PaaS) and software-as-a-service (SaaS). In general, higher degrees of multitenancy deliver greater compute efficiency and is gives the cloud its scalability and economies of scale.

On its part, virtualization is sometimes considered a specialized form of multitenancy that is implemented at the IaaS layer. Each virtual machine in the virtual environment typically runs a separate set of operating system and software, allowing for greater control and better logical separation of data – but at some expense in efficiency.

The running of multiple compute workloads on the same physical hardware is what leads businesses that are especially concerned about security to sometimes eschew all forms of multitenancy. Alternatively, they may opt to only use virtualization or multitenant software on private servers that they either own, or operate exclusively for their organizations.

Dedicated and bare metal

In contrast to a multitenant environment, a dedicated server is meant solely for running the workload of a single tenant. Unfortunately, the term “dedicated” is a much-abused one and is often used to denote anything from a standalone server to one running multiple instances with the tenant is granted exclusive control of that instance.
The loose use of this term means that a dedicated cloud offering can no longer be taken to mean that it is a single tenant environment. Enterprises where isolation and security is of utmost importance will want to clarify that they have exclusive control of the hardware server, or ensure that they are signing up for a “bare metal” offering.

Ultimately, there is a growing consensus by enterprises that a hybrid deployment works best in terms of leveraging the cost, scalability and ease of use of public clouds, and the security and service assurance of the private cloud. For more information, you may want learn how the hybrid enterprise cloud may benefit your organization here.

Follow StarHub Business on Linkedin for the latest business updates.