Ask 100 people at a cloud conference to define “cloud” and you’ll get 150 different answers. It’s a standard industry joke. So, at the risk of adding definition 151, how do we answer the question “what is cloud computing”?
In the simplest terms, cloud computing is the delivery of IT, over a network, in the form of a service. The network could be an internal company network, but when you’re talking about the cloud, it’s more usually the Internet.
IT is a pretty broad category, so immediately we must delve a bit further, at which point we encounter three bits of industry jargon you’ve probably come across: IaaS, PaaS and SaaS. Let’s look at each in turn.
IaaS - Infrastructure as a Service: This is the delivery of IT infrastructure in the form of a service, over the Internet. A service provider combines individual bits of IT infrastructure (servers and storage devices) into a single pool of IT resources – a “cloud” – which you can buy as units of computing and storage power. You can use this raw IT however you want: to run a website, e-commerce application, database, and so on. You don’t need to know which specific servers or disks will provide what you need, or even where they are.
While Amazon is probably the best-known IaaS provider, there is a rapidly-growing “cloud hosting” industry offering IT infrastructure as a service.
PaaS – Platform as a Service: This refers to the delivery of a computing platform as a service, and is most commonly used for software development. PaaS providers offer tools for application development, testing, deployment, database integration and so on. This gives developers building blocks they can use to develop software without having to worry about the underlying hardware infrastructure. The Google AppEngine is a well-known example.
SaaS – Software as a Service: This is where most people interact with cloud computing. As its name suggests, it involves the delivery of software applications as a service. Rather than installing software on your own PC or server, you access it over the Internet via a web browser. A Facebook app is SaaS. So are Microsoft Office 365 and Salesforce.com.
There are more comprehensive and technical explanations just a Google away, but in essence we’re talking about accessing different kinds of IT over the Internet.
“Hold on,” you might reasonably think. “Isn’t that just… the Internet?” Well, yes, at face value it is. Your servers, disks, data and applications live in one place, and you access them from another. Nothing new there.
The real difference with the cloud, however, lies in the way cloud-based IT works, how it is priced, and the different ways you can use it. To finish, let’s look at these three differences.
How it works: This is the real difference from traditional computing. In the cloud, you no longer talk in terms of physical computers. There is just a pool of hardware resources powering your website or applications. If you need more resources, you just grab more of the pool. When business is quiet, you can scale back your usage. And, if any individual IT component fails, the others take up the slack.
How it is priced and paid for: Instead of buying entire servers or software applications, you can pay only for the resources you need, on a utility basis. There are also economies of scale that providers can realise at the infrastructure level, by pooling their hardware resources, that tend to reduce the overall cost of cloud IT to end users.
How you access it: With cloud computing, the traditional link between applications, data and devices has been broken. IT is no longer tied to a specific server in a specific location: it’s presented to you as a service, and you can access the service from pretty much any Internet-capable device. It’s a much more flexible model that supports more flexible working methods. It gives businesses more choice, not just in the devices they use, but in where their applications and data live.
Next time, we’ll look at some of the different types of cloud, what they mean for your business, and how to choose between them.