With the emergence of cloud-computing technology comes the addition of new terminology. The terms public cloud and private cloud are reasonably well understood. However, a hybrid cloud can mean different things to different people.
Perhaps the best way to define or explain a hybrid cloud is to deconstruct the phrase into its two components — “hybrid” and “cloud.” A hybrid is, in simple terms, a combination of one or more components or ingredients. Hybrid crops combine the properties of one or more crop species to produce a crop that has disease resistance or higher yields or some other attributes. Better yet, think of hybrid cars. A hybrid car has two sources of power: a traditional gasoline engine, as well as an electric motor and rechargeable batteries. A hybrid car is a very useful example to explain the features of a hybrid cloud, and I’ll return to it later.
(Image Source – Hybrid Car – Gas2.org)
So then how to best define the cloud. The cloud is a computing environment that looks, feels, and behaves like the Internet. That’s because a cloud-computing environment utilizes standard Internet communications protocols and user interfaces to permit a variety of devices and client-side operating systems to establish a connection and thereby use the software and data resources available in a particular cloud.
Now let’s talk about deployment configurations. Sounds technical, right? It really isn’t. A cloud environment may be public; in this case, the cloud is “hosted” or installed on servers that are accessed by multiple different organizations. Each organization has its own applications and data residing in that cloud. Amazon, IBM, HP, and other companies offer public cloud environments. It is important to understand that these public cloud environments are “virtual,” meaning that they are not defined or constrained by physical limitations. Computing resources are dynamically allocated and shared across a network of servers, so that the demand for computing resources is balanced, and each individual user connection experience remains constant. The dynamic load balancing is performed by the cloud service provider and requires no action on the part of the end user. Access and data security, as well as software upgrades, are also provided by the cloud service provider, with little or no action necessary on the part of the end user.
The massive IT infrastructure of public cloud providers — in terms of servers, communication networks and bandwidth — enables them to provide cloud access from virtually anywhere in the world on a 24/7 availability basis, making them ideal for companies with global operations, as well as companies which want or need to provide 24/7 customer self-service applications. A public cloud environment may also be ideal for small and medium-size enterprises (SMEs) that need sophisticated software to run and grow their business but cannot afford the necessary investments in IT infrastructure or resources.
In a private cloud, an individual organization or a third-party provides the IT computing resources, in terms of hardware, networks, and the resource needed to maintain them. A private cloud requires IT resources to install and maintain network security, applications, and database software. For these reasons, larger organizations that already have installed robust IT infrastructure or third-party Service Providers with many clients can afford the investments necessary to build and maintain a private cloud. A private cloud has the option to offer 24/7 availability to their users and customers or not, depending upon the needs of the particular business and its available IT resources.
How is a hybrid cloud different from public or private cloud configurations? A hybrid cloud is a combination of a private and public cloud. In a hybrid cloud, some applications and data are available via a service provider in the public cloud, while other applications and data are available only in a private cloud maintained by the company or a third party. That company or third party must set up and maintain the security, applications, and databases of the private portion of the cloud configuration. In a hybrid cloud, there are most likely interfaces between the applications and data in the private and public cloud components such that it is transparent to the end user whether they are operating in the public or private portion of the cloud configuration.
Let’s return to the example of the hybrid car for a moment. A hybrid car monitors operating conditions and adjusts the power source accordingly. If the car is cruising on electric battery power and senses a demand for more power, or if the batteries are running low, the gasoline engine is engaged until the demand is satisfied or the batteries are sufficiently recharged. To ensure adequate end-user response time and the processing of transactions on a timely basis, a hybrid cloud should ideally have the ability to balance computing workloads dynamically. One way of doing this is “cloud bursting.” In cloud bursting, a private cloud “bursts” into a public cloud when additional computing capacity is needed. The additional capacity is only paid for when used.
Given the additional complexities of security, capacity, and workload balancing, why would an organization choose a hybrid cloud deployment model? The reasons are many and varied. A hybrid cloud provides an organization with existing IT infrastructure a means to extend their computing capabilities without significant new investments. Organizations with hybrid clouds can provide 24/7 customer or employee access without establishing costly off shift operations. A hybrid cloud also offers a model in which sensitive data and applications can be kept and managed separately.
Think of a hybrid cloud as a marriage of a private cloud and a public cloud. A good marriage requires effort to maintain and grow, but that effort delivers benefits that exceed the effort.
Like the hybrid crops, a hybrid cloud provides benefits via utilization of the latest technical capabilities. Like the hybrid car, a hybrid cloud can monitor the computing environment and adjust capacity and performance dynamically. Thus, for many organizations today, a hybrid cloud is indeed a marriage made in heaven — or at least, in the clouds.
By Jon Roskill