By Steven J. Vaughan-Nichols
Oct. 21, 2016
Let’s look closer. With on-demand self-service, users can unilaterally provision or access computing resources as needed. Usually, but not always, you do this with a web browser. Users spinning up a service with ordinary provisioning shouldn’t require any technical support handholding. If a technician has to manually spin up a server for you, you’re not using cloud computing. If you need to call the vendor to get a server instance up, you’re not on the cloud.
By broad network access, NIST doesn’t just mean that that cloud services must be available over the internet. It’s just that the cloud resources must be made available over the network for all devices, from PCs to smartphones, using open standard protocols such as TCP/IP, HTTP, HTML, XML, Java and SOAP. If it needs proprietary network standards or clients, you’re moving away from the standard open cloud to a proprietary solution.
With resource pooling, according to NIST, “The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter). Examples of resources include storage, processing, memory, and network bandwidth.”
Ignore the jargon. It means the cloud could be next door, or it might be in the next country. With a hybrid cloud, which uses both private and public cloud resources, it may be both. IT should know the specifics of what’s where. For the ordinary Joes and Janes in accounting, the resources are just in the cloud. From their seats, the cloud is just at their fingertips, the same way the internet is.
Rapid elasticity and expansion are vital. In a cloud, you don’t ask for five more servers; you go out and get them. Your computing resources are dynamically assigned, released and reassigned at your request. In the best clouds, users don’t even know they’re asking for more resources. They just get on with their job, and if their work requires more resources, the cloud simply provides them.
So, for example, if you suddenly need two dozen extra processors to handle an unexpected job, the cloud can deliver those compute resources to your application without any manual intervention. Then, when the job is done, those resources should automatically be returned to the cloud. No fuss, no muss.
Finally, just as with any ordinary utility, such as your electricity, on a cloud you must be able to monitor your cloud systems usage and be billed for it according to your use of the “service.” The only difference is that instead of kilowatt hours, you’re billed for storage, processing and bandwidth usage, and possibly for active user accounts.