The cloud is one of those technologies where definitions can be, well, cloudy. In previous “How to Converse in Cloud” articles, we’ve clarified and contrasted cloud deployment types and services. In this entry, we’ll tackle a common point of confusion: A lot of organizations think they’re running a private cloud, when what they’re really running is just a virtualized data center.
“Aren’t they the same thing?” you may ask. If we dig into these terms, you’ll see that they are not.
Virtualization refers to a software abstraction of workloads from the underlying physical server hardware, usually via a native hypervisor, which is a program that allows multiple operating systems to share a single hardware host. VMWare ESX/ESXI, Microsoft Hyper-V and Citrix Xen are common hypervisor solutions.
With virtualization, the operating system and applications are decoupled from the underlying hardware and containerized in a software-based virtual machine, allowing IT organizations to run multiple isolated workloads with the same or different operating systems on the same physical server. Storage virtualization abstracts physical storage, allowing diverse physical storage resources to be combined into a single virtual pool that can be sliced, diced and assigned to applications at will. Virtualization reduces the number of physical servers you need to handle multiple workloads, saving businesses costs on hardware, floor space, and power and cooling.
What makes virtualization “cloud-like” is the abstraction that enables IT to spin up, scale down and migrate workloads across physical servers in seconds or minutes, rather than the days or weeks it used to take in the physical world. In doing so, virtualization delivers much of the agility, scalability, automation, business continuity and flexibility users most often associate with the cloud. Here is where the similarity between virtualization and private cloud ends.
“Help Yourself” Defines Private Cloud
A virtualized data center is not a private cloud for one essential reason: IT is in full control and manages the entire process. Whenever a business unit, department or user needs virtual resources, it needs to go to IT to provision or scale them.
A private cloud is defined by self-service. IT or a private cloud service provides users with a large pool of virtual resources they can leverage without assistance. If a department needs a new application, it provisions the necessary virtual resources through a relatively simple self-service portal. In many cases, it can simply choose an application from the service catalog and the cloud will provision all the necessary resources automatically. When it needs more processing or storage resources, the user can simply scale them on demand, all without IT intervention. Private clouds also integrate chargeback and reporting features that let IT track and bill departments for resources used in a similar pay-as-you-go fashion as a public cloud.
Private clouds also have tremendous agility benefits for business users and departments that previously had to wait weeks or months for the IT resources needed to deploy an application or service. Before the cloud, this posed serious problems, especially in a competitive, changing business environment dependent on technology.
Of course, private clouds haven’t eliminated all resource allocation issues. With the advent of public cloud services, private clouds are being challenged by “shadow IT,” where business users work outside IT to break free from private cloud constraints and get more immediate access to cloud services without any involvement from IT or private cloud managed services.
In these cases, IT organizations need to weigh the downside of the lack of control and security shadow IT poses with the upside of having a more satisfied and productive user base. We look forward to digging into that meaty topic in a future post.