When talking about cloud computing, we will usually find two perspectives: those who believe in the concept of “private cloud computing”, and those who don’t. The people that don’t believe in the private cloud think that only on the public cloud can the benefits of cloud computing – unlimited scalability, pay-per-use, increased scalability – be fully realized. On the other side, those who believe in the private cloud feel that the potential risks of the public cloud outweigh its benefits, but that their own internal infrastructure can benefit from being more cloud-like.
The problem lies in the fact that both sides are right in their own way. There are, in fact, some benefits that the public cloud offers that cannot be replicated in a private environment. Scalability, for instance, is limited by the total available hardware, and will, for most companies, be smaller than if they were relying on external providers. It is also true, however, that for some applications and in some situations there are complex risks associated with the public cloud. There are still some questions regarding data protection and privacy that are somewhat unclear, and that makes some companies hesitant when looking at the cloud.
Enter the hybrid cloud
The hybrid cloud has arisen as middle ground between these two points of view, trying to combine the benefits of public cloud offerings while attempting to avoid the risks that companies see associated with it. The idea behind a “hybrid” cloud is exactly what the name implies: a mix between in-house and public infrastructure. From the NIST definition of cloud computing:
The cloud infrastructure is a composition of two or more distinct cloud infrastructures (private, community, or public) that remain unique entities, but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load balancing between clouds).
There are a few of extremely important details on this definition. First we have the fact that the “clouds” that are combined to make the hybrid deployment remain unique and distinct entities. This is important because it allows, for instance, companies to store their sensitive data on a private cloud, without ever exposing it to the outside, while still employing external resources to run applications that rely on this data.
Second, we have the fact that the hybrid cloud is necessarily composed of multiple cloud infrastructures. It isn’t enough to simply connect a server in your data center to some cloud resources to claim you have a hybrid cloud deployment. The private side of the hybrid cloud must operate in a cloud-like fashion, otherwise it isn’t a real hybrid cloud.
Finally, the private and public side of the cloud infrastructures needs to be linked in a way that allows the private infrastructure to take advantage, when necessary, of the resources of the public cloud to run tasks or store data. This is perhaps the most important aspect of any hybrid deployment: if your private infrastructure can’t take advantage of the capabilities of the public cloud, then you don’t really have a hybrid deployment. In fact, it doesn’t even make sense to go hybrid unless it is exactly to extend the benefits of the public cloud into your private infrastructure.
Meeting in the middle
The hybrid cloud, then, is about reaching a middle ground: about companies recognizing the benefits and the potential of the cloud, and looking to exploit it; and about cloud providers recognizing that, especially for large companies, a move to the cloud has to take into account existing infrastructure and other restrictions that may apply.
It’s also interesting to see the shift in the discourse and mentality displayed by major cloud providers in this respect. Most of them now have launched or are about to launch “virtual private cloud” services, which are the basis for the secure connection of private and public cloud environments, in order to further entice enterprise customers.