The more things change, the more they stay the same. This is true of computers, too. Some of the most cutting-edge ideas in technology have roots in the beginnings of the computer age.
Cloud computing, for instance, might seem like a recent innovation, but its basic idea has been with us from the beginning. Dumb terminals were some of the first networked computer terminals – they were capable of displaying, sending and receiving text, but had no processing power on their own. Dumb terminals depended on their connection to other computers to run programs and processes. These computers were large and expensive. Therefore, they were located in a central facility.
With the advent of PCs, users now had “intelligent” terminals by comparison: they could run programs without needing to connect to a network of machines. This brought about the age of client-server computing. This was the status quo for many years until two things happened at about the same time. The birth of the internet gave the world the opportunity to access very remote resources and data. People could suddenly access websites from all over the world. Second, virtualization provided the ability to create “Virtual Servers” running on much larger physical servers. Resources could now be allocated from the physical server to specific virtual servers in an easy and flexible manner. Cloud followed these developments.
We’ve come full circle, with cloud computing. Terminals on the cloud depend on resources shared by a group of users; these resources are dynamically allocated as needed. As software gets more resource-hungry, and as users demand more mobile data that isn’t physically tied down to a single machine, the state of the art has circled back to the networked computer model. You can’t keep a good idea down.
But cloud computing comes in a few flavors that the first users of dumb terminals could only dream about.
How do people use cloud computing in the first place? We’re all accustomed to thinking about the cloud as a storage medium: moving, storing and retrieving data over the Internet. You buy a certain amount of storage for a given time, upload your data and stop thinking about it till you need it, as the data center takes care of the nitty gritty details.
When most people think of cloud, they think of public cloud, where the cloud services can be accessed by anyone over the public internet. Absolutely anyone can set up an account on Amazon Web Services (AWS) for example and create their own Virtual Servers. Public clouds are extremely efficient and in some cases very easy to use, but it’s somewhat challenging as far as security goes. Google Drive and DropBox are two other common examples of a public cloud. All of the cloud provider’s customers share the same underlying physical infrastructure and buy pieces of it which are allocated through virtualization.
On the other hand, a private cloud involves some level of dedicated physical hardware. At one end of the spectrum are dedicated physical servers which leverage parts of a public cloud to provide network or storage resources. On the other end of the spectrum are private clouds which are entirely dedicated to the consumer of the cloud resources.
Private clouds are safer to use and easier to control than public clouds. But you pay additional costs for the dedicated physical infrastructure as the advantages of leverage and economies of scale are lost or diminished.
The hybrid cloud, as the name implies, combines services from multiple cloud providers and from both pubic and private cloud serves. As the number of cloud offerings proliferate companies are finding advantages in putting together a cloud ecosystem using a number of cloud-based solutions to build a very powerful cloud-based IT landscape.
Hybrid cloud offer far more customizability, allowing clients to pick and choose providers and services. The only real downside is that it can be a lot to keep track of, especially if different parts have to communicate with one another. Anybody who has run into a problem with too many documents in too many different places understands this. Designing, implementing and managing a hybrid cloud environment can be a challenge.
In many ways, the development in cloud has been an evolution. Like many evolutionary changes; however, the cumulative effect is revolutionary. The revolution is that in cloud, as opposed to the old dumb terminals — mainframe days, the expensive IT resources are owned, housed and operated by a cloud provider. This offloads significant work and liability from the consumer of these resources. It also improves dramatically the amount of leverage and the economies of scale which are possible. Accessibility and availability of the resources is much better in the cloud model.